Physics Quantum

This watermark does not appear in the registered version - http://www.clicktoconvert.
com
LESSION 1: INTRODUCTION TO QUANTUM

MECHANICS
CONTENTS
1.0. Aims and objectives
1.1. Quantum mechanics: an introduction
1.2. Difficulties with classical physics
1.3. The birth of quantum mechanics
1.4. State vectors
1.5. Hilbert space
1.6. Dirac notation
1.7. Norm and scalar product
1.8. Dynamical variables as operators
1.9. Change of basis
1.10. Let us sum up
1.11. Lesson end activities
1.12. References
1.0. Aims and Objectives:
In this unit, we are going to discuss

The inadequacy of classical mechanics
Birth of quantum mechanics
Basic formalism of quantum mechanics
Introduction to quantum mechanical operators
1.1. Quantum Mechanics: An Introduction
The concepts and formulation of quantum mechanics are not elementary in the
sense that they are easily understood. They are based on the outcome of considerable
theoretical research supported by experimental evidence. The quantum mechanical
approach to physical problems cannot be explained in simple non- mathematical terms.
Simple and elegant mathematical techniques have been used to elucidate the physical
concepts.
This watermark does not appear in the registered version - http://www.clicktoconvert.com
1.2. Difficulties with Classical Physics
Classical Physics deals primarily with macroscopic phenomena. Most of the

effects with which classical theory is concerned are either directly observable or can
be made observable with relatively instruments. There is a close link between the
world of classical physics and the world of sensory perception.
During the decades of 20th century, Physicists turned their attention to the
study of atomic systems, which are inherently inaccessible to direct observation. It
soon became clear that the concepts and methods of classical macroscopic physics
could not be applied directly to atomic phenomena.
1.3. The Birth of Quantum Mechanics
The early development of atomic theory consisted of efforts to overcome these

difficulties by modifying the laws of Classical Physics. These efforts reached their
successful conclusion in the period from 1925 to 1930, when an entirely new
theoretical discipline, Quantum Mechanics was developed by Schrdinger,
Heisenberg, Dirac and others.
Quantum Mechanics can be regarded as the fundamental theory of atomic

phenomena. The experimental data on which it is based are derived from physical
events that lie almost entirely beyond the range of human perception. It is not
surprising, that the theory embodies physical concepts that are foreign to common
daily experience.
1.4. State Vectors

The values of Schrdinger wave functions (x) at various points x are akin to
the components of a vector. While the components are numbers which depend on the
choice of coordinate system or basis in the vector space, the vector itself is an abstract
geometrical object with an existence and significance independent of coordinate
systems. In the quantum mechanics, the abstract mathematical object which is
associated with the physical entity, the quantum state, is known as the State Vector.
The different kinds of wave functions for a given state (in the configuration space,
momentum space) are simply the sets of components of the state vector as referred to
different coordinate systems or bases in the state vector space. This notation was
introduced by Dirac. It is used to relate different representations. The notation is very
general and suitable for handling of systems which have no classical counterpart, e.g.
spin angular momentum.
1.5. Hilbert Space
It is an abstract space. The vector spaces of quantum mechanics are called

Hibert Space. The mathematical concept of a Hilbert Space, named after the
German mathematician David Hilbert, generalizes the notion of Euclidean space in a
way that extends methods of vector algebra from the two-dimensional plane and
three-dimensional space to infinite-dimensional spaces. In more formal terms, a
Hilbert space is an inner product space an abstract vector space in which distances
and angles can be measured which is " complete", meaning that if a sequence of
vectors approaches a limit, then that limit is guaranteed to be in the space as well.
1.6. Dirac Notation
Bra-ket notation is the standard notation for describing quantum states in the
theory of quantum mechanics. It can also be used to denote abstract vectors and
linear functionals in pure mathematics. It is so called because the inner product (or
dot product) of two states is denoted by a bracket, consisting of a left part, called the
Bra, and a right part, called the Ket. The notation was invented by Paul Dirac, and is
also known as Dirac notation.
Dirac introudced the symbol y to denote an abstract state vector distinct
from its representation, the wave functions. The state vectors constitute a complex
vectors space, i.e. if y , c are state vectors, so that c y +c 'c for any complex
numbers c and c'. all vectors c y , c 0, which have the same direction y ,
correspond to the same physical state. The set of all such vectors constitute a space
called a ray in the space. Therefore, we can say that any physical state is represented
by a ray (rather than a single vector) in the vector space.
Corresponding to every vector y there is a conjugate vector y for which
Dirac used to notation y .
i.e. y y , y y (1)
The relation between any vector and its conjugate may be visualized readily if
we consider examples of their representations. In the Schrdinger picture
y y(x y x (*
) , y ). It is to be mentioned that the relation between y and
y is not a linear.
1.7. Norm and Scalar Product
The length or norm of the vector y is denoted by y y . By definition
y y is a real non-negative number.
i.e. y y 0y ,y
= 0 if and only if y = 0 . (2a)
In general if the vector y and its conjugate f of some vector f are juxtaposed in
such a was as to form a closed bracket f y , this symbol represents a single (real or
complex) number which is defined to be the scalar product of f and y , taken in
that order. Vectors of type y which forms the first half of the closed brackets are
called Bra Vectors and those of the type y which forms the second half are called
Ket Vectors. The conjugate of a bra vector is a ket vector, and vice versa.
In the Schrdinger picture, f y f* ( x )yx( d)t . It is to be noted that

*
f y y f but f y = y f (2b)
The scalar product of two vectors is linear in the second vector.
f ( c y + cc' )=f c y' + c f c (2c)
Any vector space (in general infinite dimensional) such that between any two
vectors of the space a scalar product which satisfies equation (2) is called a Hibert
space. The space of ket vectors (or bra vectors) representing quantum mechanical
states is thus a Hilbert space.
1.8. Dynamical variables as Operators
Each dynamical variable A (x,p) is represented in quantum mechanics by a

linear operator A = A ( $x , p ) -A( x,
h i ) The
. operator acts on the wave function of the
system. The effect of an operator A on a wave function is to convert it into another

wave function denoted by A . Linearity of the operator means that a linear
combinations of A 1 and A 2. is
A ( 1yc +1 c 2y) =c2 ( A1y + )1 c 2y( A 2 )
dynamic variables in quantum mechanics do not commute with each other that is
AB BA.
1.9. Change of Basis
If the vectors e i are on orthonormal basis for a N dimensional space,
( e , ( ) e,= d1
i j
, i 2= , j. . . N. , ) ,
i j (1)
Then a new basis can be constructed by forming N linear combinations of these

vectors,
N
ei =
uj *j
e
i j (2)
j =1
in which ui j are suitably chosen complex numbers. This new basis are also
orthonormal, provided that
(e , e ) = d
i' j'
i j (3)
If Equation (2) is substituted the above condition becomes
u i u*k ( ekj , lle )=i d j (4)

k, l
Using equation (1) we get
*
u
k
u =d
i k j k i j (5)
In other words, the matrix U = ui j must satisfy
U U = 1
It follows that, U must be non singular and
U -1 =U (6)
It is also apparent that this conditions is sufficient to insure that the vectors ei ' formed
according to equation (2) are orthonormal. Hence the necessary and sufficient
conditions for the vectors ei ' to form an orthonormal basis is that the matrix U is
unitary. Since U -1U =U-1 U= 1 , the equation (6) becomes

U U = U U
= 1 (7)
Such a unitary transformation has the effect of substituting the basis {e' } for the basis
{e } .If a vector x is represented with respect to the basis {e } by
x=
x i ei (8)
i
and with respect to basis {e' } by
' i'
x=
x i e (9)
i
Then, since these expression representatives of one and the same vector,
j
x i i
e= xi ju' j * e (10)
j i j,
hence by orthogonality of e j ,
u j * xi '
x j = i (11)
i
This equation provides the connection between the components xi ' xi of the
vector with respect to the bases {e} {e' } . Since U is the unitary matrix, equation (11)
can be solved for xi' in terms of xi .
1.10. LET US SUM UP
In this unit, we have studied

The inadequacy of classical mechanics
Birth of quantum mechanics
Basic formalism of quantum mechanics
Introduction to quantum mechanical operators
Hilberts space and Dirac notation
1.11. LESSON END ACTIVITIES

1. List the inadequacy of classical mechanics
2. Briefly discuss about Hilberts space
3. Give an account on Dirac notation
4. What are conditions under which an operator can be represented by a diagonal
matrix.?
5. What are bra and ket vectors?
6. Define a linear operator in Hilbert space.
1.12. References
1.A text book of Quantum Mechanics P.M. Mathews and K. Venkatesan. Tata
McGraw Hill Publishers.
2.Quantum Mechanics Leonard I. Schiff, McGraw Hill International edition.
3.Quantum Mechanics V. Devanathan Narosa Publishing house.
4.Quantum Mechanics S.L. Gupta, V. Kumar, H.V. Sharma, R.C. Sharma. Jai
Prakash Nath & Co Ltd.
5.Quantum Mechanics Satya Prakash Sultan Chand Publishers .
6.Angular Momentum Techniques in Quantum Mechanics V. Devanathan

Kluwer Academic Publishers.
7.Quantum Mechanics John L. Powell and Bernd Crasemann Narosa
Publishing house.
LESSON - 2
EQUATION OF MOTION
2.0. Aims and objectives:

2.1. Schrdinger picture
2.2. Heisenberg picture
2.3. Interaction picture (or dirac picture)
2.4. Let us sum up
2.5. Lesson end activities
2.6. Problems
2.7. Do it yourself
2.8. References
2.0. Aims and Objectives:
In this lesson we are going to study

Schrodinger picture
Heinsenbergs picture
Interaction picture
2.1. Schrdinger Picture

In the Schrdinger picture, the state vectors are assumed to be time dependent
and the dynamical variables are assumed to be time independent. The wave equation
in the Schrdinger picture is
d
ih at s ( ) =as
H t( ) (1)
d t
The subscripts refers to the ket as viewed is Schrdinger picture. T h e t o t a l

time derivative in used here, since the dependence of ket on co-ordinate or other
variable does not appear explicitly.
The Hermitian adjoint wave equation will be
10
d
-i h at s =( ) tas ( H)
d t
= a s (t ) H (2)
Since H = H because H is hermitian. Assuming H to be independent of time

the solution for the above equations will be of the form
H e
-i
a s ( t ) = eas h
( 0) and (3)
H e
-i
a s ( t ) =as (e0 ) h (4)
This can be easily verified by differentiating equation (3) and getting the equation (1)
-i H t
d i -
a s ( t ) = H as h
e (t )
d t h
d
ih at s ( ) =as
H t( )
d t
-i H t
h
The operator e is an infinite sum of powers of H , each of which is a
dynamical variable, that can be expressed as an operator or square matrix. It is
-i H t
h
essential to remember that the operator e is not an unitary transformation in the
ordinary sense of rotation of axes in Hilberts space as U, V, W. But it represents the
change of in to some other ket a s ( 0 ) . Thus the series as a whole is also a dynamical
variable. The time rate of change of matrix element of a dynamical variable W s in the
Schrdinger picture can be found out as
d d dW
a s (t W) s b (t s =) a{ s(t }W
) b (t
s s ) +a t(s ) b+ss t( s) aW t( s ) b s t{( ) }
d t d t t d t
-1 1 W s
= a s (t W) s bH ( s )+t a (s t ) +bs t( s ) at W( s H) b s ( t )
ih t h i
W 1
= a s (t ) s+s bt ( )s t aW
( s ) H[ b, s ](t )
t i h
The first term on the right hand side arises due to any explicit time dependence
W s may have. The second term on the right hand side arises due to the change in the
state vectors with time.
11
A special case of interest will b when W s is time independent, and when W s

commutes with H. In such a case right hand side will be zero and all the matrix
elements of W s are constant in time. Such a dynamical variable is said to be a
constant of the motion.
2.2. Heisenberg Picture

In Heisenberg Picture, the state vectors are assumed to be independent of time
and the dynamical variables are time dependent. We know that the time rate of change
of matrix element in Schrdinger picture is given by the equation.
d W 1
a s (t W) s b t( s )= at( s ) s t+bs( ) t s aW
( )[
H
s b
, s ] (t )
d t t i h
-i H t i H t
with a s (t =) eas (h 0 ) and a s (t )=as ( e0 ) h
substituting for ket and bra vectors
i H -t i H t i - H t
d h h h
h i W
s
H t
as ( e0 W s) b (e =0s )ae ( s 0 e )b s ( 0 )
d t t
i H t i H - t
1
( e0h s- H
+ a sW ) (b
W
H
s
h
) e (s 0 )
ih
-i H t
h
Assuming H to commute with e and also choosing the state vectors in
Heisenberg picture to be time independent as
i H t
aH ( 0= as) es h )
(= 0a (t )
since a H (t ) does not depend on time, the explicit mention of t may be omitted. The
dynamical variable W H and its derivative to be
i H T i -H T
h h
W H =e s We and
i H T i -H T
W W
H
=e h esh
t t
i H t - i H t i -H t i H t
d W
H 1 h h h h
aH W H b = H +
aH HW b e H e- Has HW s e b eH
d t t i h
12
d W 1
a H W Hb = a H H + bH
H HWa [H ,bHH]
d t t i h
d
since the state vectors in Heisenbergs picture are independent of time can be
d t
brought inside the ket a H in the above equations and we will have
d WH H W 1
aH b
H =Ha H+ b H Wa H [ H ]
,bH
d t t i h
This equation is true for arbitrary bra and ket vectors and hence we can write
d W H HW 1
= + W[ H ,H]
d t t ih
This equation is called equation of motion in Heisenbergs picture. This is

similar to the classical equation of motion for the dynamical variable F given by
d F F
= + { F , H}
d t t
where { F , H} is poisson bracket.
So the transition from classical to quantum mechanical picture can be made by

replacing the poisson bracket with commutator bracket divided by ih
2.3. Interaction Picture (or Dirac Picture)
We know that the state vectors alone or the dynamical variable alone are not
central to describe the development of a physical system with time. Both the state
vectors and the dynamical variables can be altered by means of the unitary
i oH s t
h
transformation operator e .
The quantities that are not altered are the matrix element of the dynamical
variable and this provides the physical content of the system since eigenvalues,
expectation values, and transition probabilities can be obtained from the matrix
element.
In the interaction picture, the Hamiltonian is split up into two parts as H = H0

+ H' where H0 is usually chosen to be independent of time. H0 may be simple
13
structure as the coloumb interaction and H' may be the external electromagnetic field
varying with time. In the interaction picture we define the state vectors and the
dynamical variables as
- i oH s t
h
a I (t =) a
e ( s t ) (1)
i oH st io- Hs t
W I (t =) es h W eh
Differentiating equation (1)
i H t i H t
d i o s
do s
ih at I ( ) i= h oH a hes + h (t ) sh ai s e (t )
d t h d t
i oH s t
= H o -a ts( )+ Ie a H h
s s (t )
i oH s t
= H o -a ts( )+ IH a e s
h
s (t )
= H o -a s t( )+I H
as I t( )
= H ' a I (t )
i H t- i H t i H- t i H t -i H t i Hs t
d WI i o s o s
o i W s o s o
s ho
= Ho Wh e s +he s e+h s s e Wh eo -h H s e
d h t t h
W 1
= I
+ ( I Ho- s Ho Ws I )
W
t ih
W 1
= I
+ [ I , Ho s]
W
t ih
W 1
= I
+ [ I , Ho I]
W
t ih
In interaction picture, part of the time dependence is assigned to the state

vectors and another part to the dynamical variables instead of having one constant and
the other time dependent.
2.4. LET US SUM UP

In this lesson we have discussed
Schrodinger picture
Interaction picture
in detail.
14

1. Obtain the equation of motion in Schrodingers picture
2. Obtain the equation of motion in Heinsenbergs picture
3. Compare the equation of motion in Schrodingers picture with that of

4. Obtain the equation of motion in Diracs picture
2.7. Problems:
Show that If the Hamiltonain H of a system does not contain
time explicitly
i H t

h
a (t ) = e
Solution
t
d
a ( t ) = a(t0 )+ add '' t ( )t '
0
t
1
= a ( 0 )+ a 'H t( )d' t
ih 0
t t
1 d
= a ( 0) + ' aH ' +t( ) t'a d ' (t ) ' d 't
ih 0 d 0 t ' '
t'
1 1
= a ( 0) ( )d ' t
+ 0 H a t'+ d' t ' H ( ' t)
a
th ih 0
1 1 2 2
= 1 +H. t .+ H2 . +t . a0( )
ih 2ih ! ( )
- i H t

h
=e a (0)
- i H t

h
since H is hermitian U is unitary, where U is e
2.8. Do it yourself
15
Obtain the equation of motion for <A> in the Schrodingers picture

and also in the Heisenbergs picture.
2.9. References
1. A text book of Quantum Mechanics P.M. Mathews and K. Venkatesan. Tata
2. Quantum Mechanics Leonard I. Schiff, McGraw Hill International edition.
3. Quantum Mechanics V. Devanathan Narosa Publishing house.
LESSON 3
MATRIX MECHANICS

In this lesson, we are going to solve the linear harmonic oscillator by the
method of matrix mechanics. Also we are going to study the Kronig Penny model.
3.2. One Dimensional Linear Harmonic Oscillator in Matrix

Mechanics
The Linear Harmonic oscillator can be treated by manipulating the matrix

equations directly the Hamiltonian for a linear harmonic oscillator is
p2 1 2
H= k+ x (1)
2m 2
k
where =
m
The commutator of x with H is
p 2 1
[ x, ] , x + x
H= , k x2
2m 2
1
=
2m
{ p [ x, ]p+ [ x, p] +}p0 since [x,x] = 0
16
1
= {ihp} + {ihp}
2m
2ihp
=
2m
ihp
= (2)
m
The commutator of p with H
p 2 1
[ p, ]H= , p + p , k x2
2m 2
1
=0+ k { x[ p, ]x+ [ p
, x] }x since [p,p] = 0
2
1
= k {- ihx - ihx}
2
-2ih k x
=
2
= - ihkx (3)
3.3. Energy Representations
If there are transition from the initial state l to the final state k
ih p
k [ ,x ]H = l k l
m
ih
= k p l (4)
m
But k [x , H] l =k x H -H x l= k x H
- l k H x l
Introducing a dummy operators, l l and k k
= k x l -l H l k H k k x l
= ( El - Ek ) k x l (5)
From equations (4) and (5)
ih
El - Ek k= x =l
k p l
m
ih k p l
k x l= (6)
m lE- kE
Similarly
k [ p , H] l =k i- kh x= l h-i k k x l (7)
But
k [p , H ] l =k p H -H p l= k p H - l k H p l
= ( El - Ek ) k< | p >| l (8)
17
From equations (7) and (8)

-ih k
k p= l k x l (9)
( El - Ek )
-i h k h i k p l
k p l=
( El - E k ) m -(l E k E )
h2k 1
= 2
k p l
m ( El - Ek )
h 2w 2
k p= l k 2
p l
( El - Ek )
( El - Ek =) hw
Thus we see that the energy difference between successive levels will be hw .
Therefore the energy change can takes place in units of hw
3.4. Raising and Lowering operator
Multiplying equation (6) by -ihw then we get

hw k p l
k -i mw x= l
( El - Ek)
(10)
Rewriting equation (9) as
-ihw 2m
k p= l k x l
( El - Ek)
hw
= - k i mw x l (11)
( El - Ek)
(10) + (11)
hw
k p- i w m =x l k w-p i m x l
( El - Ek)
( El - Ek kh-
w) p i -wm 0x = l
m x lis not equal to zero only when Ek = El - hw

We see that k p - i w
Therefore the operator ( p - imwx ) on l decreases the eigenvalue by hw .
Thus each operation on l will successively reduce the eigenvalue by units of hw .

18
But this cannot be done indefinitely.

Thus ( p - i) w
m 0x= 0
Now operating ( p + imwx ) from the left we get
( p +i w )(
m x -p wi m )=x0 0
{p 2
+i mw2( x - p 2 pw
) +x 2m x =0 } 0
Dividing throughout by 2m
p 2 i 2wm 2 mw 2
x
+ ih + =0 0
2m 2m 2m
p2 1 1 2
- hw + mw x 2= 0 0
2m 2 2
p2 1 1
+ m w 2 x-2 wh = 0 0
2m 2 2
1
H - hw =0 0
2
1
H 0 = h0w
2
1
Thus hw is the eigenvalue of the operator H, corresponding to the eigenket 0 .
2
We can also similarly prove that ( p + i w
m )x increases the eigenvalue by hw . Since
this is the lowest eigenvalue and each operation of ( p + imwx ) increases the eigen
value by hw , we can write
1
En = n + hw n=0,1,2,..
2
( p + imwx ) is called the raising operator or creating operator and ( p - imwx ) is called
the lowering operator or annihilation operator for the above mentioned reasons.
It is a customary practice to represent the raising and lowering operators as

dimensionless quantity. Therefore we multiply the raising operator ( p + imwx ) by the
-i i
factor and the lowering operator ( p - imwx ) by the factor
2mhw 2mhw
The creation and annihilation operator is defined as
19
1
a = m x {- wi p }
2mhw
1
a = m x {+ wi p }
2mhw
1
a
a= m x{ +wi p m}w{- x i p}
2 mh w
1
=
2 mh w
{ {
m 2wx i+2wm p(-2 x x +2)p p }
(
1 1 2 2 i wm- hi) p 2
= mw x + +
hw 2 2 m 2 m
1 1
= H + hw
hw 2
H 1
= +
hw 2
1
hw a -haw = H (12)
2
1
a a= m x{ +wi p m}w{- x i p}
2 mh w
1
=
2 m hw
{
m 2wx i +2 w
m x(-2 p p) +2 x p }
1 1 2 2 i wm p2
= m +w x +
hw 2 2 m 2 m
1 1
= -
(H h) w
hw 2
H 1
= -
hw 2
1
H = haw +a hw (13)
2
Post multiply by a in (12)
1
H =ahw a - ahw a a (14)
2
Premultiply by a in(13)
1
a =Hhw a + ahw a a (15)
2
Since aH - Ha = [a, H ] we have
[a, H ] = hwa
Similarly
20
1
a H= hwa -a hwa a
2
1
H =a hwa +ahwa a
2

Since a H- H =a a , H we have
a , H= h-wa
a, a =a a- a a
H 1
a a= +
hw 2
H 1
a a= -
hw 2
a, a =1
This is called the number operator
Eigen values of raising and lowering operator a, a
Let a y n =an yn
a y n =bn yn +1
where a n and b n are the eigenvalues,
Let us find
H 1
y n a ayn = yn - yn
hw 2
(n +1 / h2w ) 1
= yn - yn
hw 2
= n y n yn
=n (16)

y n a yn a =yna ayn
= a n+ an yn-1 yn-1
2
= an (17)
Comparing equations (16) and (17)
an = n
a y n =n yn -1
In a similar manner let us find
H 1
y n a ayn = yn + yn
hw 2
H 1
= y n + yn
hw 2
= yn
( n + 1 +/ )y2hw 1
n
hw 2
21
= ( n +1) y n yn
= n+1 (18)

y n a yn a =yna ayn
= b n bn y
+1n a y
+
n1
2
= bn (19)
Comparing equations (18) and (19)
bn = n +1
a y n =n y+n 1 +1
3.5. Matrix representation for aa, x , a n d p
y n' a yn '= ynn yn -1
= nd n' , -
n1
0 2 0 0 0

0 0 3 0 0
a =
0 0 0 4 0

0 0 0 0 5
y n' a yn =n yn +' 1 yn +1
= n + 1d n , +
n1
0 0 0 0

2 0 0 0
=
0 3 0 0
0 0 4 0

1
a = m x (- wi p)
2mhw
1
a = m x (+ wi p)
2mhw
1
a + a= m (2 w x )
2mhw
2mw
= x
h
22
h
x= a + a
2mw
{ }
0 2 0 0 0 0 0 0 0

h 0 0 3 0 0 2 0 0 0
= +
2mw 0 0

0 4 0 0 3

0 0

0
0 0 0 5 0 0 4 0
0 2 0 0

h 2 0 3 0
=
2mw 0 3 0 0

0 0 4 0
1
Similarly (a
- )a = i -( p2
2hw m
)
mhw
Or p = i - a
2
(-a)
0 -2 0 0

mhw 2 0 -3 0
=i
2 0 3 0 -4

0 0 4 0
3.6. An infinite array of square wells The Kronig Penney Model
When the array of potential wells extends to infinity on both sides, we have to
make sure that the wave functions does not diverge as x .We know that this is
possible only if m = 1 For, ,N
. if m < 1 m and m - N as N +
as N - .
When >1 the situation is reversed. In either case, the wave function in the Nth well
and N -
will diverge in one or both of the limits N + . Hence we must have
m = eiq ( real) (1)
Also we have
1
( )
c oq =s mc o m s-1+ha ' .w c =bo s w. s as+i nw h 'b=.f sw i nb -c wo s j ( )
2
(2)
where
23
1
2 2 2 2
f =( c oa s+s h w' . sai n wh ' ) (3)
t aj n= s. at a w n h '
The above equation (3) determines the allowed energy levels since and
(and hence f and also) depend on the parameter E. From the structure of this
equation it is immediately evident that the allowed energies form continuous bands
with gaps in between. For, since the value of the left hand side of equation (2) is
restricted to -1 c oq s 1 the
, equation can be satisfied only when the right hand side
also is within this range. As the parameter E is varied , the right hand side,
f c ob s w-( j) as, oscillates between the bounds + f (E) and - f (E) which lie outside
the limits +1 and -1 admissible for cos, since f c o as> w

h ' . 1The intervals of E
for which f c ob sw
- ( j ) 1constitute the allowed energy bands, and those for
which f c ob sw
- ( j >) 1are the forbidden regions. These form energy gaps,
between allowed bands. The variation of f c ob sw- ( j )with E, and the identification
of energy bands and gaps are illustrated in Figure 3.1
V =0
(n-1)' n n' (n+1)'

=
V = -V0
E ' b E
-.25 -.25 V0
V0
Fig.1.1. An array of identical square wells

-.5 -.5
-.75 -.75
-2 -1 0 1 2 -2 -1 0 1 2
Fig.1 fcos(-) b) fcos(-)
Fig.3.1 Energy bands and gaps or V0w 2 = 2 a0w0 =( ) 'w0 . 1 and ;

. 2 Bands of allowed energies are shown as thickened
(b )w ' =0 w
sections on the energy axis. Note the narrowing of the bands as '
increases. In the limit w ' they reduce to points at the positions of
the energy levels of a single potential well, shown by arrows in the
24
The nature of the energy eigenfunctions u(x) corresponding to any energy E within an
allowed band can now be easily seen. We recall that we have already imposed the
necessary continuity conditions on the wave function, which resulted in the transfer
equation
( ) An+
An-
=Mn
A-
A+
n
( ) -1
-1
(4)
and also the condition of finiteness at infinity which gave us the restriction equation
(2) on the eigenvalues of the transfer matrix M. There are no further conditions on the
wave function, and in particular, there is nothing to prevent us from choosing the two
coefficients (A+, A- or B+, B-) in any one of the regions in Figure.1.2. Suppose then
that we choose some energy E within an allowed band, and after finding ,,c,s for
this E using
2m 2m 2m 2
2
= a 2 a-2 n (+d 2 E0)= bV 0 (-V )=
E (5)
h h h
determine from equation (2) and the matrix M itself from equation (4). Now let us
avail ourselves of the freedom of choosing the coefficients, and take Ae+ , Ae- to be such
that the column made up of these is the eigenvector belonging to the eigenvalue eiq of
M, i.e.
M ( ) e= ( )
A0+
A0-
q
A0+iq
A0-
q
(6)
The -dependence of the coefficients so defined is explicitly indicated here. From
( )
A+N
AN-
= MA- N
A+
( ) 0
0
(7)
we have
25
( )
An+
An-
q
A+
= eAi- qn ( )
q
0
0
(8)
i.e. the coefficients depend on n only through the simple factor ein . It is obvious that
a similar result holds good for Bn+ a, n n-d Balso. If we think of a well and the adjacent
barrier together, as forming a unit cell of the repeating pattern, the result just
obtained may be stated in the following instructive form: The wave functions at
corresponding points x and x + nb of wells (or barriers) separated by n unit cells differ
only by the constant factor ei qn :
bq (ei qn u ) x
uq ( x + n) = (9)
This property is often expressed in the form
uq ( x =) eqiq x/ (vb x) and vq ( x =qv) x +( n )b (10)
Equation (10) may be thought of as the definition of vq ( x ,) and when it is introduced
in equation (9), the periodicity of vq ( x )follows. Besides uq ( x )there is another
independent solution u-q ( x )obtained by using the eigenvalue e - iq of M instead of
eiq in equation (6). The most general eigenfunction for a given energy E is a linear
combination c1 qu( x2+) qc -(u x), with arbitrary complex coefficients c1 and c2 .
The above results are of considerable physical importance in the study of

crystalline solids where in the array of atoms produces a potential forming a regular
repeating pattern. In the usual situation, the number of atoms is so large that the
idealization to infinite number may be made to simplify matters. The Kronig-Penny
model makes the further idealization of treating the potential due to each atom as a
square well. But the fundamental specifically, the occurrence of allowed energy bands
and forbidden gaps follows just from the periodicity property, V(x) = V(x + nb), of
the potential. The general form equation (10) of the eigenfunctions, are called Bloch
wave functions. But the explicit structure of the function vq ( x )in equation (10)
depends on the actual shape of the potential function.
3.7. LET US SUM UP

26
From this lesson one can get an clear idea on linear harmonic oscillator in detail. We
can calculate the energy eigenvalues. The Kronig Penny model, is discussed in
detail which find application field of nuclear physics and material science.
3.8. Test Yourself
1. What are raisnig and lowering operators? Explain why they are so called.
2. Give the matrix theory of harmonic oscillator and obtain the energy eigenvalue
spectrum.
3. Explain Kronig Penny model.
4. Write down the matrices for the ladder operators a,a+ of the harmonic oscillator in
the representation which diagonalises the Hamiltonian.
3.9. References
4. Quantum Mechanics S.L. Gupta, V. Kumar, H.V. Sharma, R.C. Sharma. Jai
Prakash Nath & Co Ltd.
5. Quantum Mechanics Satya Prakash Sultan Chand Publishers .
6. Angular Momentum Techniques in Quantum Mechanics V. Devanathan
7. Quantum Mechanics John L. Powell and Bernd Crasemann Narosa
Publishing house.
.
27
LESSON 4
STATIONARY PERTURBATION THEORY
4.1 Aim and Objectives
In this lesson, we are going to give an introduction to various approximation methods
and going to study Stationary perturbation theory and its applications in detail
4.2. INTRODUCTION
In quantum mechanics, there are many physical systems for which, the
equation of motion cannot be solved exactly. So, one has to resort to approximation
methods. There are three most important approximation methods in quantum
mechanics. They are
4.2.1. The Perturbation Method

This method is applicable, when the Hamiltonian can be written as sum of two
parts. i.e.
H = Ho + H '
Where H O unperturbed Hamiltonian

& H ' perturbed Hamiltonian & is assumed to be very small.
Applications : - Stark effect, Zeeman effect
4.2.2. Variational Method

It gives good estimate of the ground state energy by choosing an appropriate
trial function. The main disadvantage of this method is that, Pt cannot be extended
higher excited states.
Applications :- Ground state of Hydrogen, Helium.
28
4.2.3. W.K.B. Approximation

This method gives an approximate but direct solution of Schrdinger equation.
This method is applicable. When the potentials are such that, the Schrdinger
equation is separable to a one dimensional equation and when the potential is a
smoothly varying function.
Applications :- Barrier penetration problem, Alpha decay problem.
Approximation methods, therefore, plays an important role in all applications
of theory. This enhances rather than, diminishes, the importance of those problems for
which exact solutions can be found out. Also, the exact solutions are often useful as
starting points for approximate calculations. They may also help to establish limits of
validity for various approximation methods. These methods can be broadly classified
in to two, according as they deal with bound states or with scattering states.
4.3. Time Independent Perturbation Theory or Stationary

Perturbation Theory
The stationary perturbation theory is concerned with finding the changes in the
discrete energy levels and eigenfunctions of a system when a small disturbance is
applied. It is assumed that the Hamiltonian H is the Schrdinger equation can be
written as sum of two parts.
i.e. H = Ho + H ' (1)
where H o is unperturbed Hamiltonian & is of sufficiently simple structure,

that its Schrdinger equation can be solved.
& H ' is perturbed Hamiltonian & is small enough so that it can be regarded
as a perturbation on H o .
Let U k & Ek be the orthonormal eigenfunctions & eigenvalues of the
unperturbed Hamiltonian H o . Let y &W be the wave function and energy level of
perturbed Hamiltonian Therefore we can write.
H 0 Uk = Ek U
k
Hy =y
W
29
4.3.1. Non degenerate case
The assumption that H ' is small suggests that, we expand the perturbed
eigenfunction & eigenvalues as power series in H ' . This is easily accomplished in-
terms of a parameter l, such that, the zero, first, second.. powers of l correspond
to zero, first, second .. orders of perturbations. We replace H ' by l H ' and
express Y and W as power series in l. We assume that, these two series are
continuous analytic function of l for l between zero and one. The different orders of
perturbation approximation are then given by the coefficients of corresponding
powers of l. In the final result, l is set equal to one.
The perturbed wave function & energy level are written as
y = y 0 l+ y1 +l 2 2y +3l3 y+ . . . . . . . (3a)
W= W
0 +
l W l 2W
1 + 2 +l3 W
3 + ........ (3b)
using (2) and (3) in equation (1) we get
2 2
( H o' +)l ( .+y
H0. . l
+. 1 . Wy
. + . 2 . l =. W
)0 y( + . 1l. W
.+ l
. 2 . +. . . . )
2
(4)
(y 0. +l. y
1. +. l 2. y+
. . . . )
The above equation is valid for a continuous range of l . We now, equate the
coefficients of equal powers of l on both sides to obtain a series of equations that
represent successively higher orders of perturbation.
l 0 Hyo =0 W y
0 0
( H o- )W
0 y0= 0 (5a)
l 1 H oy +1H 0 =y 0 W1 + y1W
'W 0
( H 0- ) W0 y
=( W
1 -' 1 y
H )0 (5b)
s i m i l al r2 l
y f oo- r (0yH W
2= ) 1-
( W 1yH ' 2+) Wy
0
(5c)
l3 ( o-H W
0y )3= ( 1-W y
2 H 2+
' )y
1 W 2+ W y
0
(5d)
........................................
.........................
........................................
.........................
The first of the above set of equations means that y 0 is any one of the unperturbed
eigenfunctions. Therefore, we can write,
30
y 0 u m= & o W
= mE (6)
This state u m is discrete, since we are dealing with the perturbation of a bound state. It
is assumed to be non degenerate as well, although, others of the unperturbed
eigenfunctions may be degenerate or continuous.
4.3.2. First order Perturbation
The first order perturbation equation is given by

0 - W0 y) (1 = W 1-
( H H ' 0y)
Expressing y1 =n a n
( 1 )
u n y &0 = mu
( H
o -W0 )an= (u1 )(n w- 1H ' m) u
n
( 1 ) ( 1
n an H 0 u- n a n )n u
W 0 = n u 1 -Hm '
W m u
n a(n E 1 n u-
) (
n a n
1
E n) u m = n u 1 -H m '
W m u
( 1 )
n a [ En -E m] u=n W1 u- mH 'm u
n
Multiplying by - & rearranging we get

( 1 )
n a - n] u+
n[ Em E k W1 u= mH 'm u
Multiplying the above equation u*k and integrating we get,

(
n a [n E m 1E- ]n u) tk u dn * W
+ 1 u * k ut dm * u =k H 'm tu d
*
y.l Ty dl t1 mi= f dml =; 0 =l i f m
W . K
m
n a n( 1 [m)- E
n dk E n + ]1 dkW m =Hk m'
Carrying out the summation we get

a k ( 1[)m E-k E+
1 ]kd W
m+ k =
H m ' (7)
w h e r ke =Hm ' *uktH m' u= d k |< H ' |>m
When k = m equation (7) becomes

W1 =Hm ' m m
= | H< ' | >m (8)
which represents the first order perturbation of the energy.
31
For k m, equation (7) gives

H k m'
a (k 1 =) (9)
( mE- kE )
N o t i c e t h a t w e h a v e n o t b e e n a b l e t o d e t e r m(ni n eHa1e. n c )e w em( a s s 1u m e) t h =a t a 0
4.3.3. Second order perturbation

The second order perturbation equation is given by
( -oH W
y
0 ) =2 ( W
-1 Hy
1 ' )+
2 Wy
0
=n s ( iny2 n 1) g= ( na 1
E x p r ye2 s un ) , ny
a u0 =&m u
n
weget
n a (n H 20 u-n) a n( Wn2 u) = na(
0 n
1

W n
)
u 1 -(a n 1Hn ) ' n un+ W
2 m u
n a (n E 2n u-n) a n( En2 um) = an( 1 ) (
n n u 1 -a
W n
1
Hn) ' n un+ W
2 u
m
( 2 ) ( 1 ) ( 1 )
n a [ E -Em ] n -u 2 W m=u n a n -W 1 un a
n n n H n ' n u
Rearranging we get
n a( n [ E- En ) +]n u( 2= Wm1
m
2
un ) a n -H
( 1
n '
nu n a) 1 W n u
Multiplying by u *k & integrating we obtain.
n a ( n[ E
-m 2 E n ])tku un * d+2 W*k u mtu d
= n a ( nu1 H
k
)
' n u*t n d
(
a 1 -Hm ) ' k u* tnu d
m
n
(
n a n [-m 2 nE ) dk E n( ]+1 k W)m d
2 na
'
k
(
=
H n
1
a
n -m)H m d' k n
n n
Carrying out summation we get

a k ( [m 2 E ) ( 1 )
k- 2 E+k ]md W
' (
n =n a k S Hn
1
-ka ) m H' m (10)
For m = k, we get,
W2 = n a n S( H
m
1 ) ' (
n -am
1 ) '
Hm m
' ( 1 ) '
= n
a H
n m n
n m
where prime over the summation indicates that, we have omitted the term n = m.
Substituting the value of a (n 1 )from first order perturbation equation we have,
' | Hn m'2 |
W2 = n S
( mE- nE )
where we used the relation H m ' n= ( H n m' )*. The above equation gives
the second order perturbation to the energy eigenvalue.
For k m, equation (10) gives
32
a (k [ 2
m E -)k E (n =] 1n a )kSH(n '-1k a )
m H m '
u s i n g t h e v a l u e o(n 1f )a w e g e t
H ' H ' H ' H '
a (k 2 =) n' S k n n- m k m m m2
( mE - Ek ) m(- E n E ) ( mE- kE )
< k | H> ' | n <n | H '> | m k< | H ' >| m m<| H ' >
| m
a (k 2= )
n
'
S - 2
(12)
( mE - Ek ) m(- E n E ) ( mE- kE )
Similarly we can obtain W3, W4 . . . . . . . ,

w ith1,theenergyandwavefunctiontosecondord
l = erinH'isgivenby
1
|< m | H > | 2n |
W =E mm
+ | <H '> |n m+ 1 (13 a)
E m - En
|< k | H >' | m
|<
' m | H>' | m | <k | H >
' | n | <
n | H ' >| m
y um = '
u+ k 1 k- + n (13 b)
E m - E n( E m-E )k ( m-E E k ) -(m E n E )
The primes in the above equation denote the omission of k = m or n = m. It is to be

noted that, the wave function y is not normalised since
2
|<k | H >' | m |
(y , y ) = 1 + '
k ( mE- kE )2
(14)
f t h e c o n d iyt0 i oy
tothesecondorderinH'.Thisoccuredbecauseo s n (=
, ) 0>, s 0
Applications of Stationary Perturbation Theory

The Normal Helium Atom Or Ground State Of Helium Atom
-e r12 -e
r1 r2
+Ze
NUCLEUS
Consider the normal state of helium atom which contains a nucleus of charge
Ze and two electrons each having a charge e. The potential energy of such a
system is given by
Z 2 e Z2 e 2 e
V= - - + (1)
r1 2 r 1 r2
in which r1 and r2 are the distances of electrons 1 and 2, respectively from nucleus
33
and r12 is the separation of the two electrons as shown in Figure 1.

Considering the nucleus at rest, the wave equation for the two electrons
becomes
2y 2 y 2 y 2 y 2 y 2 y
2 + 2 + 2 + 2 + 2 + 2
x1 y1 z1 2x 2y 2 z
(2)
2m Z e 2 Z 2 e 2 e
+ 2 E + + - y= 0
h r1 2 r 1 r2
The variables x1 , y1 , z1 are Cartesian coordinates of one electron and x2 , y2 ,

e2
z2 those of the other. If the term is omitted, the wave equation can be solved
r1 2
exactly; hence this term may be regarded as perturbation term
2
H =
(1 ) e
(3)
r1 2
The unperturbed wave equation can be separated into two equations
by substituting
(0 ) (0 )
y ( ) ( x1 y1 z1 , x 2 y 2 z 2 ) = u1 ( x1 y1 z1 )u 2 ( x 2 y 2 z 2 )
0
(4)
Considering the wave functions in polar coordinates r1 , 1 , f 1 , and r2 , 2 f 2

and in normal state,
y 1( 0)0 y
0 =
(0 )
, 1 0q 0( r1 , j
0
0 0 1,)y
( )
1 r1 1,(q0 ,j
0 2 2 2 )
= y 1(s0) (q1r , 1 r1,(
js0, 1) y )
(2 q, 2 ,j2 )
and corresponding eigenvalue is
(0 ) (0 ) (0 ) 2
E100 ,100 = E1 + E 2 = -2 Z E H (5)
The first order perturbed energy E(1) is the average value of the perturbation
( )1
e2
function H = over the unperturbed state of the system. Hence
r1 2
e 2 ( 0 )2
E (1) = y n (H0 ) *yn( 1= ) ( 0 )
dt y d1 0 0 , 1t0 0 (6)
r1 2
we know that
Z3
u1(s0) 1= y00 ( )0= e 3
-r / 2
(7)
p a0
in which = 2Zr / a0 and a0 = h2 / (4 2 me2 )
34
Z 3 ( - r1/ 2 ) ( - r/ 2 )
y 1( 00)0 , 1 0=0 e e . (8)
p a03
The volume element is
2 2
dt = r1 dr1 sin q 1 dq 1 df1 r2 sin q 2 dq 2 df 2 (9)
So that the integral for E(1) becomes
Z 2e p
2p p 2 p e( - r1 - r)2
E (1) = d r12 r1
25 p 2a0 0 0 0
1
0r 0 0
2 (10)
2
s q1id q1n dj-
1 s rd i r
2 ndq2
2 q
2 d 2j
in which 12 = 2Zr12 / a0.

The value of the integral can be calculated by considering the
electrostatic interaction energy of two spherically symmetrical distributions of
electricity, with density functions e(-1) and e(-2) respectively, Solving the integral,
we have
5
E ( 1 =)Z EH (11)
4
Thus the total energy of the system is
E ( =1 (
) 2
-
-2Z5/4Z ) H E. (12)
Thus we can calculate the energy of He, Li+, B e ++, etc. with Z = 2,3,4 etc.
respectively.
4.4.2. Stark effect in Hydrogen: Linear or First order Stark Effect
The ground state of hydrogen atom is non degenerate, but the n = 2 energy
levels of hydrogen atom is four fold degenerate. On the application of an external
electric field, the degeneracy is partly removed. The splitting of atomic energy levels
due to the application of external electric field is known as Stark effect and it has been
discovered as early as 1913. It is found that the splitting of energy levels of the
hydrogen atom for weak electric field is proportional to the field strength, and hence it
is known as Linear Stark Effect or First order Stark Effect.
The splitting of energy level for all other atom is proportional to second power
of the field strength and that is known as Quadratic Stark Effect.
35
Let us denote the eigenfunctions of the hydrogen atom, by y n l lm(qr , f, ) . The
ground state y 1 0 0is non degenerate, but, the first excited state y 2l m
l
with n = 2 is
four fold degenerate. We shall write down explicitly the degenerate eigenfunctions,
u1 =2yR0 =0 r2q 0( Y0) 0 f( , )

u2 =2yR1 =0 r2 q 1( Y1) 0 f( , )
(1)
u3 =2yR1 =1 r2 q 1( Y1) 1 f( , )
u4 =2yR1 - =1 r2q 1(- Y1) 1f( , )
where, Yl m
l
(q , f) are the spherical harmonics and R21(r) and R20 (r) are the
radial part of the wave functions. The radial functions are given by ,
1 r - r / 2a0
R2 (0 = ) r -1 e (2)
2a03 2a0
1 r - r / 2a0
R2 (1 =
) r e (3)
2 6a0 a03
2
a0 = h 2 / m e (4)
The Hamiltonian of the hydrogen atom is taken as the unperturbed
Hamiltonian H0 and the applied electric field as the perturbation H'. If the electric
field E is along the z axis, the perturbation is given by
H' = - eEz
= - e- Er cos
4p
H 'e= Er 1 0 q jY(,) (5)
3
where, - e- is a positive quantity.
Since, H' is of odd parity, the matrix elements between the states of same
parity (same l value) vanish. The same result can be arrived at by considering the
angular integration involving the spherical harmonics. The only surviving matrix
elements are,
y2 0 H0y' 2 1 0

y2 1 H0y' 2 0 0
It can be easily seen that this two matrix elements are equal. Let us
evaluate one of them
36
y2 0 H0'y 2 1H =0*y2 d0 0 y
' 2r W1 0 d
= R2 (r 0 )R 2( r 21) r '0 dW( 0 r ,q1H) f0 Y( q Y

, f )d
4p
= e E 2 (R 0 ) r 2 ( R13) r1( Wr,0q d)0 f (r0 q, Y1 f) Y0 (q Y, f )d
3
1
where, Y00 (q , f =
)
4p
Y Y W
d =1
1 0 1 0
1
and hence the angular integration yields . The radial integral can be performed
4p
by substituting the radial function from equations (2) and (3)
-r
3 1 r 4 a0
R ( r) R( ) rr
0
2 0 2 1
0
=d r -1r e
4 3 2a0
d r
r
Letting x = the integral becomes,
a0

a0 x 4 - x
1 - x e d x
4 30 2
=
a0 5 !
= 4 -!
4 3 2
9a
= - 0 (7)
4 3
4p 91a
y 2 0 H0y' 2 1 0 e= . E - . 0
3 3 4p
4p
= e E
3
y2 H
0 y' 0 e 2 =
3 1E- 0 a0 (8)
Denoting the matrix elements y 2 0 H0y' 2 1 0 and y 2 1 H0y' 2 0 0 a s H1' 2 and
H 2' because
1 of the definition equation (1) and denoting the quantity 3|e|E a0 by P1 we
have,
H1' =H' 2= P-2 =
e 1-
3E 0 a (9)
Substituting the values of H1' 2and H 2' 1in the secular equation,
37
-W1 - P 0 0
- P -W 1 0 0
=0
0 0 -W1 0
0 0 0 -W1
The roots of the above secular equations are ,

W1 =P +
, P -, 0 , 0
Let us write down explicitly the set of linear equations

-W1 C-1 P 2 C= 0
- P 1C- 1W 2C= 0
(10)
- W1 C3= 0
- W1 C4= 0
If W1 = P, then C1 = - C2 and C3 = C4 = 0
If W1 = -P, then C1 = C2 and C3 = C4 = 0
If W1 = 0, then C1 = C2 = 0 and C3 and C4 are indeterminate
Then it follows that,
1
U1 = u1 ( - 2u )
2
1
U 2 = u1 ( + 2u ) (11)
2
U 3 =C3 u3+ C 4 4u

U 4 =C3 ' u3+ C
'
4 4u
give respectively the first order correction to the energy and the corresponding
zeroth order eigenfunctions. It is seem that out of the four degenerate states only the
two states y 2 0a 0y
n 2 d1 0 are split but the other two states y 2 1 and
1 y 2 -1 continue
1 to be
degenerate. This is diagrammatically shown in the following figure.
U1 W0 +3_e|Ea0
W 0 u1 u2 u3 u4
U3 , U4 W0
U2
W0 -3_e|Ea0
in absence of electric field in presence of electric field

38
4.5. LET US SUM UP

It is very clear that this lesson gives an exposure on stationary perturbation
theory and its application such as normal state of Helium and Stark effect.

Check your progress
1. There is no first order stark effect in the ground state of an atom Illustrate.
2. Explain briefly stationary perturbation theory.
3. Define the degeneracy of an energy eigenvalue and illustrate its removal by
considering the stark effect in the n = 2 level of the hydrogen atom.
4. Obtain the expressions for the first order corrections to the energy and wave
function making the use of the non-degenerate stationary perturbation theory.
4.7. References
2. Quantum Mechanics A.K. Ghatak and S. Lokanathan McMillan Publishers.
3. Quantum Mechanics Satya Prakash Sultan Chand Publishers
5. Quantum Mechanics S.L. Gupta, V. Kumar, H.V. Sharma, R.C. Sharma.
Jai Prakash Nath & Co Ltd.
39
LESSON 5
VARIATIONAL METHOD AND WKB
APPROXIMATION
5.1. Aims and Objectives

In this lesson also, we are going to concentrate on the other two approximation
methods namely Variational method and WKB approximation.
5.2. Variational Method:

The variation method is specially applicable for determining the lowest energy
of the system. According to this method, the integral
W = f* Hf td E0 (1)
is an upper limit to the lowest energy state of the system. In the above
equation, H is the complete Hamiltonian operator for the system and the function f is
any normalised function of the co-ordinates of the system. The choice of the function
f is quite arbitrary but the more wisely H is chosen, the more closely W will
approach the energy Eo . If we use the true wave function y o of the lowest state instead
of f , then W will be equal to E o .
It f is not equal to y o , we may expand f is terms of the complete set of
normalised orthogonal functions y o , y
, 1. . y2. . y
.n .
f a= w iyt n n n h *na= a 1 (2)

n n
Substitutinginequation(1)weget
* *
W = a ya H y
n m n dm z
n m
* *
=
ay n a
m y n dm
E
m z
n m
*
=
a n dam mE n m
n m
= a *n an En
n
2
W = a n En (3)
n
Subtractingthegroundstateenergyfromthebo
thsides
40
2 2
W- E
0 =a n E-
n E Q a n =1
0
n n
2
= a n (- n E 0E ) (4)
n
2
As a n is positive and E n E0 , the R.H.S is positive or Zero. Thus the have proved
that W is an upper bound to the lowest value of energy E 0 .
QW E0 (5)
Note :-
If the function f is not normalised, then W is given by the expression
*
W=
f Hf dt
*
f fd t
The variation of f is performed by first choosing a suitable functional form for f .
Depending on a number of parameters l1 , . .l2. . . .ln . . . . .
. . ) f1* ( l, 2. .l. .n l
W (l1, . l. .2 .l. . n= . . . 1y) H l(2 , .l.2 . l. . )td
As the parameters are varied, the value of W also varies. Minimization of W is then
accomplished by finding the values of l1 , .l2. . ln. . for which
W W W
= . .=. . . . . . . .=. . . . . . . 0
1 l 2 l n l
5.3. Application of Varitional Method :

5.3.1. Ground state energy of the Hydrogen atom.
In this case wave function is of the general form
y = e- a r
(1)
For the ground state of hydrogen atom, the potential energy term will be

e2 / r and hence the form of the Hamiltonian operator H will be
h2 2 e2
H =- 2 - (2)
8p m r
as the energy occurs in the radial equation, we need only to consider the radial part
of the Laplacian operator 2 . Hence
1 2
2 = r (3)
r 2 r r
The value of 2y will be

41
1 -a r
2y = 2 2 r e
r r r
2
1 2 - ar
= 2 2r + r e
r r r 2
2
1 - ar 2
= 2 r e + r ( )e - ar ( )
r2 r r 2

1
=
r 2
[
2r (- a )e - ar + r 2 a 2 e - ar ]
2a
= e - a ra-2 (4)
r
According to variational method, the energy E is given by

E=
y * Hy dt
y * yd t

-a rh 2 -a 2 a 2r 2 - ea 2 r
0
e - a8p 2 m e- e - r
r r
d 4p r

=
e -2 a r4r p2 d r
0
where 4pr 2 dr is the volume of a spherical shell at a distance r and r+dr from the
nucleus. The limit of integration varies from 0 to as r goes form zero to infinity.
h2 2
2a - a
h2 r 2 2-a2a r 2 - a 2 r
- r e +d r r e - d r e r e d r
0 8p 2 m 08p2 m 0
=
(5)
2 -2a r
r e d r 0
n!
Applying the relation x n e - ax dx =
0 a n +1
The values of the above integrals can be obtained as
h2a2 2 h2a 1 2 1
- + -e 2
8p 2 m 8a 3 4p 2 m 4a 2 4a
=
2
3
8a
h2a2
= 2
- e2a (6)
8p m
Now we shall choose a in such manner as to give a minimum energy. Thus
E 2h 2
= -a e2= 0
a 8pm2
42
4p 2 me 2
Or a= (7)
h2
Substituting the value of a in equation (6), we have
2
h 2 4p 2 me 2 2 2
2 4p me
E min = -e
8p 2 m h 2 h
2

2p 2 m 4 e
Em i n= - 2 (8)
h
This is same value as we have obtained by other methods.
5.4. The W.K.B. Approximation

5.4.1. Introduction
In the development of Quantum Mechanics, the Bohr- Sommerfield
quantization rules of the old Quantum theory, occupy a position intermediate between
classical & Quantum Mechanics. It is interesting that, there is an approximate
treatment of the Schrdinger wave equation that shows its connection with the
quantization rules. It is based on an expansion of wave function in powers of h ,
which has asymptotic nature is nevertheless useful for the approximate solution of
quantum mechanical problems in appropriate cases. This method is called the
Wentzel kramer Brillouin or WKB approximation. It is applicable to potentials
which are such that, the Schrdinger equation is separable to one dimensional
equations. Further, the potential, should be slowly varying such that
1 d k
< k<
k(x)dx
1 / 2
1 2 m
where K(x) = p x =( E2 V
- ( x ) represents the local wave number. It is of interest
h h
to mention that the motion of a particle in a varying potential is analogous to the

behaviour of light in a medium of varying refractive index. If the refractive index
changes rapidly, one has reflections. If the change is gradual, the trajectory is not
straight, but there are practically no reflections. The behaviour of particle in a varying
potential is quite similar.
This is the approximate method of solution of wave function based on the
expansion, of which, the first term leads to classical results, the second term to the old
43
quantum theory results and the higher terms, to the characteristics of the new
mechanics.
One dimensional Schrdinger equation is written as
-h2 2 y
+ ( -V y
E )= 0 (1)
2 mx 2
The WKB approximation takes advantage of the fact that the wavelength is changing
slowly, by assuming that the wave function is not changed much from the form it
would take of V were constant, namely
( i 2p x )
y e wh =h e r e=2 p 2 m- ( E V )
This suggests that, when E > V, it will be convenient to write, the wave function in
the form
( i S )
y =e h (2)
where S is a function of x (S may be complex).
It V is nearly constant, we may expect that s is roughly equal to p 2 x. Now, S can be
expanded as
h2
+h(1 x ) S+ (2 x ) . +. . . . . . . . . . .
S =S (o x ) S (3)
2
Using equation (2) and (3)

yi S
y e(= ih S/ )e =( h i S / )
x h x
2 i y i( ih SS /2 ) S( h i S i / ) S
2 = e e+
x h h x x2hx
-1 ( S i
i h S / ) 2 h( i S / )S2
= e (+ ) e
h2 x h 2
x
equation(1)becomes
1 S 2 S 2 ( i Sh / )
- ih ( 2 + V E- ) =e 0
2 m x x

2
1 S S 2
- ih (+ V - E )= 0
2 m d x x 2
2
1 S i hS 2
+ (- V - E )=2 0 (4)
2 m d x m x
SubstitutingthevalueofSfromequation(3),( h
uptosecondorderin)inequation(4)weget,
44
2
1 So S h 2S
+ h 1+ 2
2 m x x 2 x

ih 2 S h 2S 22 S
+( - V - 2oE h+ 2 1 )+ 2 2 =
0
2 x m x 2 x
2 2
1 So S S S 2SoS
+ h 2 1 + o2 h h12 + x
2 m
x
x
x
x x
i 2So 2S1
2
+( V E
- ) -h +h 2 =0
2 m x 2 x
2
1 So S0 i h S 2
+ (- V E +) -0 1
2 m
x m x x 2 x 2
2
(5)
h2 So S2 S1 2S
2
+ + - i = 0
2 m x x x x 2
Since this equation must be satisfied, independently to the value of h , it is necessary

that, the coefficient of each power of h be separately equal to zero. This requirement
leads to the following series of equations.
2
1 So
+ - ( V =E ) 0
m (6)
2 x
o 2
1 So Si1 S
- = 0
m x x 2 x 2
S S Si 2
( o o r1 - o ) 2= 0 (7)
x x 2 x
1 So S2 S1 2 1S 2
+ - i = 0 ( o r )
2 m x x x x2
So S2 1S 2 S1
- -i 2 =0 (8)
x x x x
Fromequation(6)wehave,
So 1 2
= [ 2m (-E v] )
x
x
S0 [= 2 m ( E- V ) 1 ] 2 d x (9)
x0
From equation (7) we have

S1 i 1 2 So
=
x 2S ( o x/ ) x 2
i S
= l o g o
2 x x
i S
S1 = l o g o (10)
2 x
45
-1 / 2
-1 S S
i 1S= l o = go l oo g
2 x x
1 1
o r i 1eS = = 1 4
S o / x [ 2m (E- V)]
similarlyfromequation(8)wecanshowthat
1 m ( v / x ) 1 m 2( v /2 x )
(11)
2 [ 2 m -( E2 v3 ) ] 2[ -2 m5 ( E v2 ) ]
S2 = -d x
NeglectingS,thesolnofequation(1)isgivenb
2 y
i i 0S
( oS+ h S1 )
i 1S
y e h= e =h e
Asshaspositiveaswellasnegativevalues,
0
y A =( i0eSh / ) +
i 1
e S
B ( -ii 1S oh / S)
e e
i x
A +
= -e ( [ 2 m ( E V )1 ]2 d x )
h
[( 2 m ) -( E V] 1 )4 x0
i x
B -
+ -e (h [ 2 m ( E V )1 ]2 d x )
[( 2 m ) -( E V] )
1 4
x0
x x
A i ( p2 Bd x i ) 2( p- d x )
= e h x0
+ eh x0
(12)
p2 p2
where A & B are two arbitrary constants and P2 =2 m (- E V . )The positive exponential
represents a wave moving in the positive X direction while the negative exponential, a
wave moving is the negative X direction. When V is a constant, the exponential
reduces respectively to plane waves,
i i -
( 2p x ) 2( p x )
eh h
e
When V > E, the W.K.B approx possesses real exponential solution. The solution is
x x
i i -
A h x0 1 B hx0 1
p d x p d x
y =e + e
p1 p2 (13)
w h e r1 e= p 2 m- ( V E )
5.5. LET US SUM UP
By reading this lesson one can get an understanding on the two approximation
methods namely variational method and WKB approximation and its applications.
Variational method is applicable to study the ground state of system. WKB method
find its application in alpha decay paradox.
46

Check your progress
1. Explain how ground state energy could be evaluated using the variation
method.
2. Briefly discuss an application of WKB method.
3. Briefly outline the variation method.
4. Explain how the ground state energy could be evaluated using the variation
method and hence find the ground state energy of helium atom.
5. Explain the main stages involved in the WKB approximation. Use the
technique to obtain the asymptotic solution of the one dimensional
Schrdinger equation. Also arrive at the solution near a turning point.
5.7. References
1. Quantum Mechanics S.L. Gupta, V. Kumar, H.V. Sharma, R.C.
Sharma. Jai Prakash Nath & Co Ltd.
2. A text book of Quantum Mechanics P.M. Mathews and K. Venkatesan.
Tata McGraw Hill Publishers.
Publishing house.
47
LESSON 6
TIME DEPENDENT PERTURBATION THEORY
AND ITS APPLICATIONS
6.1. Aims and Objectives
We are going to discuss
Time dependent perturbation theory and its applications
Fermi Golden rule
Harmonic perturbation
Constant perturbation
Selections rules
6.2. Time Dependent Perturbation Theory
If the Hamiltonian of the system is time dependent, the energy is no longer a

constant of motion. In such case we have to solve time dependent Schrdinger
equation.
Ey H
= y
y
ih =
H y (1)
t
In many physical problems, the time dependence of the Hamiltonian arises due
to an external agency. For example, an atom irradiated by light, by applying electric
field to atomic system etc. In such cases, Hamiltonian is written as,
H =H 0 V
+ ( t ) (2)
where, H 0 is the sum of Hamiltonian of the two systems in isolation and V(t) is
their interaction.
At a particular instant of time t = 0, the system may be regarded as being
approximately in an eigenstate of H 0 as i. We have to calculate the probability of
finding the system at a later time t in an eigenstate f. If y n denotes the complete set
of eigen functions, then, we have
48
H 0 y n E=n yn
= h wn yn
The wave function corresponding to the actual state can be expanded in terms
of y n
y n ( t )= tn ) ( r )- i ewn t
nC ( y (3)
n
E - En
where, w s - n w = s is the Bohr angular frequency between the states s and n. If the
h
perturbation V(t) is absent, all the co-efficients C n will be constant in time. On the
other hand, the perturbation V(t) makes them vary with time and for this reason, this
method is called the method of variation of constants.
Suppose the perturbation begins to act at time t. Initially at t = 0, the system is
in the unperturbed state say, i before the perturbation is turned on. In such a case,
Cn ( 0=) 0
f o r n i
Cn ( 0=) 1
f o r= n i
1
Cf =Vf ( i t ) +ei wf ti
t ih
Integrating the above equation between the limits 0 and t, we get,
t
C f t 1 iw t
0 t d 0=tihV (f ti ) ef i d t
t
1
Cf ( ott =) fV (i t ) iwef ti d t
ih 0
t
1
Cf ( t-)f C (=f O ) iV ( t i)wf eti d t
ih 0
Q Cf ( 0= ) 0
t
1
Cf ( t =f) V i( t ) eiwfdti t , f o r f i
ih 0
This gives the probability amplitude for states of later time other than the
initial state.
6.3. Fermi Golden Rule (or) Closely Spaced Levels

(or) Transition to continuum
If transitions are taking place to a set of closely spaced levels f then, Pf ( t )
will be obtained by summing over, all final state levels f Therefore,
49
Pf ( t =)f | C ( t2 ) |
| fV (i 2 o f 2) wi| t2
= s i n
/i 2
f
h2 wf
If the level spacing is very close, then, the summation can be replaced by integration.
= f
r( Ef ) df E
where r ( fE )is the density of the final states around E f (i.e) the number of energy
levels per unit energy interval.
| V (2 0 ) | tw / 2
Therefore, Pf ( = t ) f r2( f Ei ) f s 2 ifi n d E
h w f /i 2
| f V 2( i 0 f) | 2 i tw / 2
= r ( fE ) s i f n d E
h2 w f /i 2
A s s u m i n g e ( fE ) d o e s n o t vfa r y a r o u n d E , w e h a v e ,
| V 2( 0 ) | t w / 2
Pf ( = t f )2 r( f i E )f s 2 iif n d E
h - wf /i 2
| V 2( i 0 f) | 2 it w / 2
Pf (= tf f) r( E f )w s ii = hn dQ [w hE ]
h2 - wf /i 2
| V 2( 0 ) | t w / 2
Pf (= t f 2)f (i rE hff) s2 i w
i in d (1)
h - w f /i 2
Since the integral is practically zero outside w f ,i the limits are chosen from - to + ,
consider, the integral,
+
2 w f ti / 2
s
-
i w n/ 2 dw f
f i
i
p u =
t xw
f ti / 2
dw f it
d =x
2

2 w+ f ti / 2 s i2 n x 2
s w w f /i - 2
i n = df i . d x
-
( x /t 2t )
+
2 2s i n2 x
t - x 2
= t d x
+
s i2 n x
= 2 x 2d x
t
-
+
s i2 n x
we can show that d =x p
-
x2
+
2 w t / 2
s i f ni d w f 2 = t p
- w f /i 2
50
substitutingtheintegralvalueonequation(1
)weget,
2
V ( 0 )
Pf ( =t f) i f (r E )p 2 t
h
2p t 2
= Vf r ( 0i ) f( E )
h
Therefore the transition probability per unit time, will be,

Pf ( t )2p
= Vf (r 0i 2 ) f( E )
t h
-
ThisiscalledtheFermiGoldenrule.
6.4. Constant Perturbation
If the perturbation last from time t = 0 to t = t, and during this time, it is

constant, then, we have,
t
1
Cf ( t=f ) Vi ( 0 )iwf eti d t
ih 0
t
1 e iwf( i t )
= Vf (i 0 )
ih i wf i 0
Vf (i 0 )iwf ti
Cf (= t -) e- 1
hwf i
Thus, the probability of finding the system, in the state f at time t is given by,
2
Pf ( t=)f C ( t )
V 2f ( i 0 )
Pf =( t -)
h 2 w 2f i
( iwf
) ( e-
e t i - 1 if w it
1 )
w h e
( iwf
) ( )
r - tei - ,i f ew -ti 1 wif =e it -
1 - 1f i w-ei t e+ 1
= 2 - -ei f w+t i e- if wi t
= 2 2-c o ws f (i t )
= 2 [ -1 c owsf (i t ) ]
w t
= 2 2 s i2 n f i
2
w t
= 4 s i2 n f i
2
2
V ( 0 ) w t
\ Pf (= t )f2 i 42 s i 2n f i
hwf i 2
2 2
Vf (i 0 ) 2 wf ti / 2
= s i n
h2 wf ti / 2
In such a case the probability amplitude will be,

51
-V ei wf( - ) i wte f wi i(+ )w 1t -

Cf ( t= ) 0 +
h w f - i w f wi + w
Thereforethetransitionprobability,
2
2 V 2 ei wf ( - )1i -wt f wi e(+i -)w t 1
Pf ( t=f) c ( t ) =02 +
h (wf - )i w f w( i + )w
We find that the terms inside the box, while squaring will not contribute
anything because when w f i approximately equal to w only one term (IInd term) will
contribute while the other (Ist term) vanishes. Similarly, when w f i approximately
equals - w , the contribution by the two terms will exchange. Thus, the above
discussion treating the terms independently is valid.
6.5. Harmonic Perturbation
We know that the probability amplitude for transition from initial state i to
the final state f will be,
t
1 iwf( i t )
Cf ( =
tf ) V i ( t ) e d t
ih 0
we introduce a harmonic perturbation given by

-i w
Vf ( i = t 0 ) w V( t )e > , 0
t
1
Cf ( = t 0 ) V - i we( t ) eiwf( i td) t
ih 0
Vo t i
= w e -(f w
) t
ih o
i
t
V ei w(f i-) w t
= o
ih i wf ( i- w) o
Vo
= [ i ( wef -)i wt1
- ]
ih w
i f i-( w)
V ei ( wf -)i w1t-
= -o
h wf i- w
This expression differs in having w f -i wfor harmonic perturbation instead of

just w f i for constant perturbation. The entire discussion of constant perturbation can
be carried for harmonic perturbation, with (w f -i ) w replacing w f .i We conclude that for
large t, only those transitions in which w f -i w 0= can take place with appreciable
52
probability In other words, only when E f - Ei = h w , the transitions will be appreciable.

Thus, the harmonic perturbation V0 -ei wt can induce transitions to a level E f whose
energy is higher than Ei by hw . Such a transition can be described as absorption of
energy hw by the system from the external agency.
On the otherhand, if the Hermitian adjoint V0 -ei wt is used for harmonic
perturbation, then we will have, (w f +i ) w instead of (w f -i ) w . In such a case, the
probability amplitude is given by,
- V ei (wf +i t) w - 1
Cf ( t= ) o
h wf + i w
This induces a transition in which

w f +i w 0=
h w f i= h - w
E f- iE=i (- h ow r )
E i - fE i= h w
In this case we find that E f is lower than Ei by hw and this describes the emission of
energy, Here, the energy is given by the system to the external agency.
f i n a l f i n a
i n i t i a l i n i t i
In reality, the actual potential has to be Hermitian and must be,

V0 (iw +te- i ew t )
6.6. Selection Rules
In many cases of interest, Vf (i tcan

) be expressed in the form,
Vf ( ti )iwfef =V
ti wif it
i ( o ) e
whereV(0)isindependentoftime
f i
\e q u a t i o n ( 4 ) b e c o m e s
53
t t
C f 1 iw t
0 t i h0V f( i o ) ef i d t
d =
t
t
1
Cf ( t=f ) Vi ( o )iwfeti d t
ih 0
This equation shows that Vf (i tis) non- zero only when Vf (i o is) non-zero, Thus,
transitions in the first order, can take place between pair state(f,i) for which Vf (i o ) .0
Transition between any other state are forbidden in the first order. This is the first
order selection rule. Even if Vf (i 0= ) then,
0 transition between the state i and f may still
be possible in the higher order via an intermediate ( l ) state. If both Vfl ( 0 and
)
Vli ( 0 ) ,0then the intermediate state l will connect the initial state i and final state f.
However, probability for transitions in the higher order will be smaller and will
proceed at a slower rate.
If we plot the probability Pf ( t as) a function of the energy difference w f i
between the initial and final state of fixed t, the graph will be of the form as shown in
the figure below,
P(t)
f
p t 3 / 5p/ t
p t 5 / p t 3 /
6p/ t 4p/ t 2p/ t O 2p/ t 4p/ t 6p/ t
w t / 2
The behaviour of sin s i n f i is depicted in the figure.
w f i/ 2
The main peak of the curve occurs at w f =

i 0 and has a high proportional to t 2 . The
subsequent peaks are much smaller and occurs at 3 p/ t , 5p etc.
t Thus, we see that
the transition from the initial state i to the final state f occurs such that w f i falls
within the main peak.
2p
| w f i| =
t
54
taking only the positive value,

2p
wf i =
t
2p
hw f i = h
t
hwf i= h t
which resemble uncertainty principle.

The magnitude of the energy difference between initial and final is very
unlikely to be significantly to be higher than h/t. This is the energy time uncertainty
relation.
6.7. LET US SUM UP
Here we have studied in detail the followings:

Time dependent perturbation theory and its applications
Fermi Golden rule
Harmonic perturbation
Constant perturbation
Selections rules
Check your progress

1. Define Fermi Golden rule.
2. What do you meant by selection ruels.
3. Obtain the expression for the transition probability per unit time for first
order transition under constant perturbation and show that it is proportional
to the perturbation time t.
6.9. Problems
Calculate the second order energy correction for the ground state of the charge
between linear harmonic oscillator subject to a constant electric field. Compare the
result with the exact solution
55
Solution
The schrodinegers equation for the perturbed system is
-h 2 d 2y 1 2
+ k y+ x a x=y E y
2m d 2 x2
where is the product of the charge and the magnitude of the electric field in the x-
direction. Writing X = x + / k, the above Equation becomes
-h 2 d y 1 2 2 a1 2
2
+ k =yX E+
2m d2 X 2 k
1 1 a2
Thus En = n + hw- , which is the exact result.
2 2 k
First order energy correction is zero.
The second order energy correction is given by
2 ( 0 ) ( 0 2)
'a n x m
En( 2 =) 2 m
(E( 0 ) ( 0 )
n -E
m )
Expressing x in terms of the raising and lowering operators,
( ( 10 0h) )( 0 0) )( a 0 )
n0 x ) =m( n a -( m n m
i 2mw
1 h
= - md n , +m dn ,+1 m 1
i 2m w
- m1
a2 a2
= En( 2 =) - = -
2mw 2 2 k
Do it yourself
1. A hydrogen atom the energy state with n = 2 is subject to an electric field E
parallel to the z axis. Calculate the energy correction.
2. Calculate the splitting of hydrogen energy levels due to spin-orbit interaction.
56
6.10. References

3. Quantum Mechanics Satya Prakash Sultan Chand Publishers
Publishing house.
57
LESSON 7
ANGULAR MOMENTUM ALGEBRA

In this lesson we are going to discuss the one of the most important and
exciting topic of quantum mechanics namely, angular momentum algebra.
7.2. Introduction
Knowledge in angular momentum techniques is essential for quantitative study
of problems in atomic physics, molecular physics, nuclear physics and solid state
physics. The application of angular momentum techniques to solve physical problems
has become so common. Angular momentum techniques are widely used in the study
of both non relativistic and relativistic problems in physics. In this chapter, we will
discuss the Quantum Mechanical definition of Angular m omentum first. Then we
will have a discussion on physical interpretation of Angular m omentum. It is then
followed by, discussion on ladder operators and coupling of two angular momenta.
7.3. Quantum Mechanical definition of Angular Momentum

In classical mechanics, the angular momentum vector is defined as the cross
u r
product of the position vector p . ie
r v u r
L = rp x (1)
v u r
Both r p& change sign under the inversion of co-ordinate system and so
r
they are called polar vector. It is easy to see that, L behaves differently and will not
change sign under inversion of co-ordinate system, and hence it is known as pseudo or
Axial vector.
u r r u r
L= r p
i j k
u r
L= x y z
px py zp
r
L = (y pz z py ) i (z px x pz) j + (x py y px ) k
58
i.e. Lx = y pz z py
Ly = z px x pz
Lz = x py y px
[Lx , Ly ] = [y pz z py , z px x pz]
= [y pz , z px ] [y pz , x pz] [z py , z px ] + [z py , x pz]
= y px [pz, z] + py x[z, pz]
= i h y px + i h p yx
= i h [x py y px ]
= i h Lz
i.e. [Lx , Ly ] = i h Lz (2)
Similarly
[Ly , = i h Lx
Lz ]
[Lz , Lx ] = i h Ly
r r
i.e. ( L x L ) = L x Ly Ly L x
r r
(LxL) = i h Lz
Expressing the angular momentum in units of h , we can write it as

r r r
(LxL) = i L (3)
This Quantum mechanical definition must be generalized to include angular
momentum of half integral values. Therefore, generalizing, we can write the
definition for total angular momentum as
( J x J ) = i J (4)
Now, [Jx , Jy ] = i h Jz
[Jy , Jz] = i h Jx (5)
[Jz , Jx ] = i h Jy
7.4. Physical Interpretation of Angular Momentum Vector

Although the components of angular momentum operator do not commute
among themselves, it is easy to show that, the square of angular momentum operator.
r2 2 2 2
J = Jx + Jy + Jz
commutes with each of its components.
i.e. [J2 , Jx ] = 0
59
[J2 , Jy ] = 0 (6)
[J2 , Jz] = 0
Equations (5) & (6) gives simple physical interpretation. It is possible to find
the simultaneous eigenvalues of J 2 and one of the components, say Jz alone, but, it is
impossible to find precisely the eigenvalues of Jx and Jy at the same time.
2
Representing the operators by matrices. One can say that, J and Jz can be
diagonolised in the same representation, but not the other components Jx and Jy .
Physically this means that, one can find the magnitude of angular momentum vector
and its projection on one of the axes. The projection on the other two axes cannot be
determined. This is illustrated in the figure(Fig 7.1) , in which the angular momentum
vector is depicted to be anywhere on the cone. If y jm is the eigenfunction of the
r
operators J 2 and Jz, then
r2
J y jm = hj y jm
& Jz y jm = m y jm
Here j & m are Quantum Numbers used to define eigenfunctions and the
corresponding eigenvalues of the operators are hj & m.
Z axis
Jz
r r
J J
Fig: 7.1 The angular momentum vector, for which its

projection on the z-axis alone is well defined, but not its
projection on the x axis & y axis.
60
Problem 7.1
Show that [J2 , Jx ] = 0
Solution
r r
We know that J2 = J x J
= Jx 2 + Jy 2 + Jz2
\ [J2 , Jx ] = [Jx 2 + Jy 2 + Jz2 , Jz]
= [Jx 2 + Jx ] + [Jy 2 + Jx ] + [Jz2 + Jx ]
= Jx [Jx , Jx ] + [Jx , Jx ] Jx + Jy [Jy , Jx ] + [Jy , Jx ]Jy
+ Jz [Jz , Jx ] + [Jz , Jx ] Jz
= i h Jy Jz i h Jz Jy + i h Jz Jy + i h Jy Jz
i.e. [J2 , Jx ] = 0
Similarly [J2 , Jy ] = 0
& [J2 , Jz] = 0
7.5. Ladder Operators (or) Raising and Lowering operators

We define the raising and lowering operators as J+ and J given by
J+ = Jx + i Jy
J = Jx i Jy
It is to be noted that, though the angular momentum matrices are Hermitian,
the ladder operators are not Hermitian.
{J+} = (J+) + (i Jy )
= J+ i Jy
= J
Similarly {J } = (J+) (i Jy )
= Jx + i Jy
= J+
\ J+ & J are not Hermitian.
Problem 7.2
Evaluate [J2 , J+]
Solution
Now, [J2 , J+] = [J2 , Jx + i Jy ]
= [J2 , Jx ] + [J2 , i Jy ]
= 0 + i (0)
61
= 0
Similarly [J2 , J ] = 0
Also [J+ , J2 ] = 0
and [J , J2 ] = 0
Problem 7.3
Find the value of [Jz , J+] and [Jz , J ]
Solution
Now, [Jz, J+] = [Jz , Jx + i Jy ]
= [Jz , Jx ] + i [Jz , Jy ]
= i h Jy + i ( i h Jx )
= i h Jy + h Jxs
= h (Jx + i Jy )
= h J+
And, [Jz, J ] = [Jz , Jx i Jy ]
= [Jz , Jx ] i [Jz , Jy ]
= i h Jy i ( i h Jx )
= i h Jy h Jx
= h (Jx iJy )
= h J
Hence we conclude that [Jz , J] = h J

Also J2 Jz = Jz J2
and J2 J+ = J+ J2
Problem 7.4
Find the values of Jz J+ and Jz J
Solution
We Know that, [Jz , J+] = J+ in units of h
i.e. Jz J+ J+ Jz = J+
or Jz J+ = J + Jz + J +
i.e. Jz J+ = J+ (Jz + 1)
Similarly [Jz , J ] = J in units of h
i.e. Jz J J Jz = J
62
Jz J = J Jz J
i.e. Jz J = J (Jz 1)
7.6. Spectrum of eigenvalues for J2 and Jz

Let us define J2 y jm = hj y jm
and Jz y jm = m y jm
The raising and lowering operators can be defined as
J+ = Jz + iJy
and J = Jx iJy
Let J+ y jm = j+
and J y jm = j
Now, Let us operate J2 on j+ and j
J2 j + = J2 J+ y jm
= J+ J2 y jm
= J+ hj y jm
= hj J+ y jm
= hj j+
i.e. J2 j + = hj j+
Similarly J2 j = J2 J y jm
= J J2 y jm
= J hj yjm
= hj J yjm
= h j j
i.e. J2 j = h j j
Operating JZ on j+ and j we have,
Jz j + = Jz J+ yjm
= J+ Jz yjm
= J+ (Jz+1) y jm
= J+ (m+1) y jm
= (m+1) J+ y jm
i.e. Jz j + = (m+1) j+
63
Similarly Jz j = Jz J y jm
= J Jz y jm
= J (Jz 1) yjm
= J (m1) y jm
= (m1) J y jm
i.e. Jz j = (m1) j
Thus we find that j+ and j are eigenfunctions of J2 operator, with the same
eigenvalue hj. However, even though we find that, j+ and j are also eigenfunctions
of Jz operator, the eigenvalues are either stepped up or stepped down by unity. It is
precisely for this reason, J+ and J operators are called as raising and lowering
operators, respectively.
Now let us find the value of (J2 x + J2 y ) y jm
(Jx 2 + Jy2 ) y jm = (J2 Jx 2 ) y jm
= (hj m2 ) y jm
Since the eigenvalues of square of the Hermitian matrix is always positive, we

have (hj m2 ) 0. Therefore the m values are bounded. ie.. m has a spectrum of
eigenvalues, the lowest being h jand the highest value being + h j.
Now, let us denote the lowest value of m by m1 , and the highest value of m by
m2 .
Obviously J+ jm = 0 (A)
2
and J jm = 0 (B)
1
Operating J on the left of equation (A) and J+ on the left of equation (B), we have
J J+ jm 2 = 0
(Jx iJy ) (Jx + iJy ) jm 2 = 0
[J 2
x ]
+ J 2y - i (J y J x - J x J y ) jm 2 = 0
(J 2
x + J 2y - J )
z jm 2
= 0
(J 2
- J 2z - J )
z jm 2
= 0
64
[J 2
]
- J z (J z + 1) jm 2 = 0
[ j - m 2 (m 2 + 1)] jm 2
= 0 (C)
Similarly J+ J jm1 = 0
[ j - m 1 ( m 1 + 1) jm1 ] = 0 (D)
From equations (C) and (D),

[ - m (m
j 2 2 + 1)] = 0 (E)
[ - m (mj 1 1 - 1)] = 0 (F)
Since R.H.S of (E) and (F) are equal to zero, we can write,
j - m 2 ( m 2 - 1) = j - m 1 ( m 1 - 1)
i.e. m 2 (m 2 - 1) = m1 (m1 - 1)
m 22 - m12 + m 2 - m1 = 0
(m1 + m 2 ) (m 2 - m1 + 1) = 0 (G)
Since (m 2 - m1 ) is always > 0, (i.e positive), the second term cannot be zero, we have
(m1 + m 2 ) = 0
m1 = - m2
Taking the highest value of m2 to be equal to j. i.e. m2 = j, we have m1 = j. Thus we

find that Jz has a spectrum of eigenvalues ranging from j to + j, in steps of unity.
i.e. j, ( j + 1) ..( j 1), j
We also find the value of hj
Substituting m = j in equation (E),
hj j (j + 1) = 0
hj = j (j +1)
Thus the eigenvalue of J2 operator is j (j +1).
Note
1. J2 y jm = j (j + 1) y jm
2. Jz y jm = m y jm
Where m = j, j +1,..(j 1), j in steps of unity.

65
7.7. Eigen values of J+ and J

Let J+ y jm = j+ = + y j,m+1 (1)
Now <j+ | j+ > = < J+ y jm | J+ y jm >
= y * jm J J+ y jm d
= y * jm (J2 Jz (Jz+1) y jm d
= y * jm[j(j+1) m (m+1)] y jm d (2)
Now, <+ y jm+1 | + y jm+1 > = y * jm+1 + { } + y jm+1 d
= y * jm+1 |+ |2 y jm+1 d
= |+ |2 y* jm+1 y jm+1 d
= |+ |2 (3)
Using equations (1), (2) and (3) we can write,
|+ |2 = j(j+1) m(m+1)
or + = +
j(j1) -+ m(m1)
= j2 + j - m 2 - m
= (j - m) (j + m + 1)
In a similar fashion, we can show that,
= (j + m) (j - m + 1)
Thus the eigenvalues of J + and J operators are (j - m) (j + m + 1) and
(j + m) (j - m + 1) respectively.
7.8. Matrix Elements
The matrix elements for J2 , Jz, J+ J are give below.

< j' m' | J2 | j m > = j (j+1) d jj'd m'm
< j' m' | Jz | j m > = m d jj'dm'm
< j' m' | J+ | j m > = (j - m) (j + m + 1) d jj'dm'm+1
< j' m' | J | j m > = (j + m) (j - m + 1) d jj'dm'm1
The above matrix elements are sufficient to construct all the angular
momentum matrices.
66
Note
1. 1
Jx = (J + + J - )
2
2. 1
Jy = (J + - J - )
2
3. Since all the matrix elements defined above connect states with the same j but
different m values, it is usual to construct angular momentum matrices for a
given j value.
4. It is customary to label the rows by m' values (i.e. m' = j..j) and the
columns by m values (i.e. m = j..j)
5. For a given j, the dimension of the angular momentum matrices are given by
(2j+1) X (2j+1)
Problem 7.5
Find the angular momentum matrices of J2 , J z , J + , J , Jx and Jy
when j = .
Solution
Dimension of the matrix
The dimension of the matrix is (2j+1) X (2j+1) i.e. (2 X +1) X
(2 X +1) = 2 X 2. Then, the possible values of ma are and .
The angular momentum matrix for J2 is given by.
m'\m 1
2
-1
2
1
2 3
4 0
-1 0 3
2 4
3 1 0
=
4 0 1
The angular momentum matrix for Jz
m'\m 1
2
-1
2
1
2 12 0
-1 0 -1
2 2
67
1 1 0 1
= = sz
2 0 - 1 2
For J+
m'\m 1
2
-1
2
1
2 0 1
-1 0 0
2
For J
m'\m 1
2
-1
2
1
2 0 0
-1 1 0
2
Angular momentum matrix for Jx

1
Jx = (J+ + J )
2
1 0 1 0 0
= +
2 0 0 1 0
1 0 1 1
= = sx
2 1 0 2
Similarly for Jy
1
Jy = (J+ + J )
2i
1 0 1 0 0
= -
2i 0 0 1 0
1 0 1
=
2i - 1 0
1 0 1
=
2 - 1 0
1 0 - i 1
= = sy
2 i 0 2
68
Note
1 0 1 1
Jx = = sx
2 1 0 2
1 0 - i 1
Jy = = sy
2 i 0 2
1 1 0 1
Jz = = sz
2 0 - 1 2
are the Paulis spin matrices for j = .
Problem 7.6
Obtain the angular momentum matrices J2 , Jz , J+ , J , Jx and Jy for j=1.
Solution
Dimension of the matrix (2j + 1) x (2j + 1)
= (2 x 1 + 1) x (2 x 1 + 1)
= 3 x 3 matrix
j = 1 possible values of m are 1, 0, 1.
Angular momentum matrix for J2
m'\m 1 0 -1
1 2 0 0 1 0 0
0 0 2 0 = 2 0 1 0

-1 0 0 2 0 0 1
The angular momentum matrix for Jz

m'\m 1 0 - 1
1 1 0 0
0 0 2 0

- 1 0 0 - 1
Angular momentum matrix for J+

m'\m 1 0 -1
1 0 2 0 0 1 0
0 0 1
0 0 2 2 = 2
- 1 0 0 0 0 0 0
Angular momentum matrix for J

69
m'
m 1 0 -1
1 0 0 0 0 0 0
0 2 0 0
= 2 1 0 0

- 1 0 2 0 0 1 0
Angular momentum matrix for Jx

1
Jy = (J+ + J )
2
0 2 0 0 0 0
1
= 0 0 2 + 2 0 0
2
0 0 0 0 2 0

0 2 0
1
= 2 0 2
2
0 2 0
0 1 0
2
= 1 0 1
2
0 1 0
0 1 0
1
= 1 0 1
2
0 1 0
Angular momentum matrix for Jy

1
Jy = (J+ + J )
2i
0 2 0 0 0 0
1
= - 2 0 2 - 2 0 0
2i
0 0 0 0 2 0

0 2 0
1
= - 2 0 2
2i
0 - 2 0
0 - 1 0
i
= 1 0 - 1
2
0 1 0
70
Problem 7.7
Construct the angular momentum matrices for j = 3/2.

Solution
3 3
Dimension of the matrix = 2x + 1 x 2x + 1 = 4 x 4 matrix
2 2
3 1 -1 3
\ possible values of m are , , , .
2 2 2 2
15 3
2 0 0 0
4 0 0 0
15 0 1
0 0 0 0 0
4 2
J2 = Jz =
0 0 15 0 0 -1
0 0
4 2
15 3
0 0 0 0 0 0
4
2
0 3 0 0 0 0 0 0

0 0 4 0 3 0 0 0
J+ = J =
0 0 0 3 0 4 0 0

0 0 0 0 0 0 3 0
0 3 0 0 0 - 3i 0 0

1 3 0 4 0 1 - 3i 0 - 2i 0
Jx = Jy =
2 0 0 0 3 2 0 2i 0 - 3i

0 0 0 0 0 0 3i 0
LET US SUM UP
By going through this chapter, one can easily understand the algebra of
angular momentum. We can calculate the spectrum of eigenvalues for J2 and J z
operators. One can easily construct the angular momentum matrices for a given j by
studying this lesson.
71
LESSON END ACTIVITIES

Check your progress
1. Give the quantum mechanical definition of angular momentum.
2. Obtain the spectrum of eigen values for J2 and Jz.
3. What are called raising and lowering operators ? Why they are called so ?
7.11. Do it Yourself
1. If J = J1 +J2 show that J X J = ihJ.
2. Express the matrix for Jx when j = .
3. Find [Jz,J+]
4. Show that if any operator commutes with Jx , Jy it also commutes with Jz
5. Show that J2 and Jz form complete set of commuting observable and j can
take all integral values and half integral values greater than one or equal to
zero.
7.12. References
4. A text book of Quantum Mechanics S.L. Kakani Kedarnath Publishers.
Quantum Mechanics Leonard I. Schiff, McGraw Hill International edition.
72
LESSON 8
COUPLING OF TWO ANGULAR MOMENTA OR
CLEBSCH GORDON (C,G) COEFFICIENT

Here we are going to study the coupling of two angular momenta in detail
8.2. Clebsch Gordon (C,G) Coefficient
r r r
When we couple two angular momenta J1 and J 2 , the final J will be between
|J1 J2 | in steps of unity. We will have two sets of mutually commuting operator
given by
Set I : J12 , J2 2 , 2 J z , J
Set II : J12 , 22J ,1 zJ2 , zJ

The state vectors or wave functions corresponding to the two sets will be
j1 j 2 j mand j1 j 2 m1 m
2
The state vector corresponding to the first set j1 j 2 j m is called the coupled
representation. And the state vector corresponding do the second set j1 j 2 m1 m

2 is
called uncoupled representation.

In order to move from coupled representation to uncoupled representation, or
vice versa, we need a unitary transformation coefficient. The transition from one
representation to another representation is accomplished by the following two
equations.
j1 j2 j
j1 j 2 m1 m
2 = m m2 m
j1 j 2 j m (1)
jm 1
j1 j2 j
j1 j 2 j m = m m2 m
j1 j 2 m1 m
2 (2)
m1 m2 1
j1 j2 j
The symbol is called the unitary transformation coefficient
m1 m2 m
or vector addition coefficient or Clebsch Gordon (C,G) Coefficient.

73
Now, taking scalar product with j1 j2 j' m' on both the sides of equation (1)
j1 j 2j' m1 ' j2 j 1m 2m
j1 2j j
j1 j 2j' m1 ' j2 j 1m 2m = m m2 m
j1 j 2j' m1 ' j2 j 1m 2m
j m 1
j1 2j j
= m d d
m2 m j 'j m 'm
j m 1
Now, doing the summation over j and m on R.H.S, because of the presence of
Kronocker delta, of all the possible values of j and m, only those terms corresponding
to j = j' and m = m' will remain.
j1 j2 j
j1 j 2j m1 j2 j 1m 2m = m (3)
1 m2 m
Thus we find the C.G. coefficient is the scalar product of the two state vectors
in the coupled and the uncoupled representation.
In a similar manner, taking the scalar product with | j1 j2 m1' m '2 > on both the
sides of equation (2),
j1 j2 j
j1 j 2 m1 m2 j1 j2 j m = m (4)
1 m2 m
Taking hermitian adjoint (Congugate transpose) of equation (4)
j1 j2 j
j1 j 2j m1 j2 j 1m 2m = m (5)
1 m2 m
8.3. Orthonormal Properties of C.G. Coefficient
We know that the wave functions in the coupled representation as well as in

the uncoupled representation are orthogonal.
j1 j 2j' m1 ' j 2j j m = d j 'j dm 'm (1)
j1 j 2 m'1 m'2 j1 j2 m1 m
2 = d m m' dm ' m (2)
1 1 2 2
74
Decoupling equation (1)

j1 j j'
j1 j2 j' m' =
2
' j1 j 2 m'1 m
'
2
m1' m2' m1' m2' m
j1 2j j
j1 j 2 j m = j1 j 2 m1 m
m1 m2 m 1

2m m
2
Substituting in equation (1), we have

j j j' j1 2j j
m1 m2 1' 2
' m j1 j 2 m'1 m2 ' j1 j'2 m1 ' m
2 = d j 'j d
m2 m
m 'm
m m2' m 1
m1' m2' 1

j j j' j1 j2 j '
m1 m2 1' 2
' m m m d m1 m' 1dm2 ' = d jj' dmm'
m m2' m 1 2
m2
m1' m2' 1
Now, summing over m1' and m '2
j1 j 2 j 1 ' j 2 j j
= d j 'j dm 'm (3)
m1 m 2 m 1 m 2m '1 'm 2 ' m m
In a similar way, starting from equation (2), the orthogonal property for the
C.G. coefficient can be written as
j1 2j j j1 j j
2
= d m m' d (4)
j m m '1 m' 2 m ' m1 m2 m 1 1 m 2 m '2
8.4. Symmetry Properties of C.G. coefficients

1. j1 j2 j j j1 j
m = (-1) j1 + j2 - j 2
1 m2 m m 2 m1 m
2. j1 j2 j j j2 j
m = (-1) j1 + j2 - j 1
1 m2 m - m1 - m2 m
3. [ j] j1 j j2
= (-1) j1 -m1
[ j2 ] m1 - m - m 2
4.
( - 1 j2 + m)2
[ j] j j2 1j
= -m
[ j1 ] m2- m
1
where [ j ] = (2j+1)1/2
75
8.5. Conditions to be satisfied to C.G. coefficient to exist
1. Triangle Rule
D (j1 j2 j) must be satisfied. i.e. the final j must be within the permissible limit
of |j1 j2 | to j1 + j2 (in steps of unity).
2. m sum rule must be satisfied
m 1 + m2 = m
3. Projection rule must be satisfied i.e. m should be valid projection of j. i.e. m = j,
j +1 .(j1), j is steps of unity.
From equations (3) and (4) we have
2
j1 j2 j
= 1 (5)
m1m 2 m1 m2 m
2
j1 j j
2
= 1 (6)
j, m m1 m2 m
8.6. Some Special Types of C.G. Coefficients
1. When either j1 or j2 is zero, then the coupled and uncoupled representation, will
become equal. In such a case, the corresponding C.G. coefficient will be equal
to unity.
0 j j
i.e. 0 m m = 1

j 0 j
m 0 m = 1

However it should be noted that, if the final j = 0, then

j j 0
m - m 0 1

76
8.7. LET US SUM UP
We have studied the coupling two angular momenta in detail in this chapter.
Also we discussed the symmetry properties of C.G. coefficient.
8.8 . LESSON END ACTIVITIES

Check your progress
1. What are C.G. Coeffifiient? Briefly explain their significance.
2. List the symmetry properties of C. G. coefficient.
3. What are the conditions to be satisfied for C.G. coefficient to exist?
4. Write down the special type of C. G. coefficient.
5. Give the orthonrmal properties of C.G. coefficient.
8.9. References
77
LESSON 9
RELATIVISTIC QUANTUM MECHANICS
In this lesson we will be studying the important aspects of relativitic quantum
mechanics such as Klein Gordon equation and Diracs relativistic wave equation.
Then we will have a discussion on the free pariticle solution for Diracs equation.
Also we are going to study the properties of Diracs matrices.
9.2. Introduction
The Quantum Mechanics developed so far, does not satisfy the requirements
of special theory of relativity as it is based on a non relativistic Hamiltonian. The
wave equation we used to discuss physical problems is of the first order in time and of
second order in space co-ordinates. This contradicts the space-time symmetry
requirement of relativity. The difficulties suggest two ways of obtaining relativistic
wave equations. i) to use a relativistic Hamiltonian for a free particle is place of non-
relativistic one ii) to try for a first order equation in both space and time co-ordinates.
Though Schrdinger himself suggested the first possibility its implications were
studied in detail by Klein and Gordon in 1926. Hence it is popularly known as Klein
Gordon (K.G.) equation. It does not say anything about the spin of the particle. Hence
it is used to describe spin zero particles such as in pions, kaons etc. Based on the
second idea, Dirac developed his relativistic wave equation. The concept of spin has
evolved automatically from Diracs theory which is applicable to spin systems.
9.3. Klein Gordan Equation
The relativistic relation between energy and momentum is given by

E2 = c2 p2 +m2 c4
and the operators for the E and P are

E = ih p = -ih
t
The Schrdinger equation will be
Hy = Ey
78

c 2 p 2 + m2c 4 y = i h y
t
But it is very difficult to give a physical meaning for the square root of an
operator. Further if the square root is expanded in terms of binomial services we will
have even powers of on the L.H.S. In order to avoid the difficulties Klein-Gordon
proposed that the wave equation will be
H2 y = E2y
2

{c 2 2
p+ 2 4
}
y c ih y
m =
t
u 2r 2
{ 2
(
c - ih 2m + 4c h=
y ) } 2
- y2
t
u 2r 2
{ - h 2 c 2m + 2 ch =y4 } -2 y 2
t
Dividing throughout by - h 2 c2
2 m2 c2 1 2
- = y y
h2 2
c 2 t
2 2
2 1 m c
-2 2
=2 - y 0
c t h
m 2c 2
{2 - } y =0 (1)
h2
where the DAlemberts operator, which is relativistically invariant
1 2
2 = 2 -
c 2 t 2
Equation (1) is relativistically invariant, if y transform like a scalar.
9.3.1. Demerits of Klein Gordon Equation

2
1. Since it involves the second order time derivative , to have a future
t 2
y
knowledge of the system we need to know not only y but also .
t
2. The product y * y in KG equation does not represent the probability current
density.
3. The wave function y in the KG equation is not orthogonal.
79
4. There is no way of introducing spin in to the KG equation and hence it is

applicable only to spin zero particles like pions and kaons.
9.3.2. Plane Wave Solutions: Charge and Current Densities

Solutions of equation
2
1 2y mc
2 2
- 2y + y =0 (1)
c t h
corresponding to particles of definite momentum p = hk may be obtained by
d2 f
substituting y ( x, t ) = f (t ) e ik . x . This leads to 2
= [c 2 +
(k2 / m
2
)h2c ] f. Solving this
d t
we obtain, the plane wave solutions
i ( k m.w
x t ) ih (p x . -E ) t
y ( x , =)t e= e (2)
1
E = hcw =p ( 2 m2 + 2) c 4 2 (3)
In contrast to the nonrelativistic case where the coefficient ( E h ) of ( it) in
the exponent is the positive quantity ( p 2 / m2h ,) here we have solutions with -w as
well as with +w. The appearance of the negative energy solutions is typical of
relativistic wave equations.
Another difference from the nonrelativistic case is that y * y cannot be
interpreted as the probability density P(x,t). We expect P(x,t) to satisfy a continuity
3
equation so that P ( x, t ) d x is time independent. To obtain such an equation we
multiply equation (1) on the left by y*, its complex conjugate equation by y, and
subtract. The resulting equation can be written as
1 2 y
2 2
y* - (y * y - yy *) = 0
c t t
This is the continuity equation with
ih y y *
P ( x, t ) = 2
y* -y (4a)
2mc t t
ih
S ( x ,t= )- y( y*- y
* y ) (4b)
2m
A convenient choice of a common constant factor in P and S has been made
here. With this choice, S coincides exactly with the corresponding nonrelativistic
expression. However, P is quite different. It vanishes identically if y is real, and in the
80
y
case of complex wave functions, P can even be made negative by choosing
t
appropriately. Clearly, P cannot be a probability density. One could multiply P by a
charge e and then interpret it as a charge density. However, P can still have different
signs at different points, and this is hardly satisfactory in the description of a single
particle of given charge. It is known that equation (1) can be used to describe a system
of arbitrary numbers of particles and their antiparticles by treating y itself as an
operator function instead of a numerical valued function. At this level, the above
mentioned properties of P are no longer objectionable.
9.4. Diracs Relativistic wave Equation
When the wave equation is of first order in time, it must be first order in space co-
ordinates too. Dirac, was probably influenced by Maxwells equation as they are first
order equations in both space & time co-ordinates. In 1928 Dirac proposed a
Hamiltonian that is linear in mass and energy given by
uurr
(
H = ca . +mpb c 2)
Dirac equation will be
u r r
Ey = { a( c.+ )pb }
y2 mc
u r r
{E -(ac .) bp 2
-y }
=
mc 0 (1)
Since we want this equation to be independent of space and time co-ordinates,

a and b must be independent of space and time co-ordinates. Otherwise this equation
will give rise to space and time dependent energies and momenta.
Further we want this equation to be linear in space and time derivatives which
already occur in b and E, a and b must be independent of space and the time
u r
derivatives. This means that a and b commute with P and E. But this does not
necessarily mean that a and b commute among themselves.

u r r
{
Premultiplying equation (1) by E + ca .+ pb 2
(
mcwe get) }
u r r r u r r
{E + (ca .+) pb 2
}{
mc E - ca( ) }
. b p - 2y mc
= 0
81
{ E+ c (a x px + ay py + a z pz) + bmc2 } {E c (a x px + a y p y + a z p z) - bmc2 }y =0
[ E2 c2 (ax 2 px2 + a y2 p y2 + a z2 pz2 ) - b2 m2 c4 c2 (a x ay + ay ax ) px py
c2 (ay az + a z ay ) pz py - c2 (a z ax + a x a z) pz px mc3 {(a x b + b

ax ) px
+ (a yb + bay ) py + (a zb + ba z) pz} ] y = 0
Comparing the above equation with the relativistic relation between energy
and momentum is
E2 c2 p2 m2 c4 = 0
we get
ax 2 = ay 2 = az2 = b2 =1
ax a y = -ay ax ax b = - b ax
ay a z = -a z ay ay b = - b ay
a z ax = -a x az a z b = - b az
Thus we find that a and b anticommute among themselves in pairs. Further

we find that eigenvalues for all the four matrices are +1 and -1
9.4.1. Properties of a and b Matrices

1. Since a and b anticommute among themselves in pairs they cannot be
numbers. Therefore they can be conveniently represented by matrices.
2. Since we want the Dirac Hamiltonian to be Hermitian, a and b must also be
Hermitian. Therefore a and b must be square matrices.
3. The eigenvalues of a and b matrices +1 and -1.

4. Since in a group of four anticommuting matrices in pairs, we can have only
one of the matrices to be the diagonal matrix we choose b matrix to be the
diagonal matrix.
5. axb = - bax
Pre multiplying by b on both sides
baxb = - b2 ax Since b 2 = 1
= - ax
Taking Trace on both sides
Tr (baxb) = - Tr(ax )
82
Tr (b 2ax ) = - Tr(ax )
Tr(a x ) = - Tr(ax )
2 Tr(a x ) =0
Tr(a x ) = 0
In a similar manner we can prove that
Tr(a y ) = 0
Tr(a z) = 0
Tr(b) = 0
6. Since the traces of all these four matrices are zero, and their eigenvalues are
+1 and -1, there should be same number of +1 along the diagonals as the
number of -1 in b matrix. Therefore the dimension of b matrix must be even.
7. The ith element of the equation
axb + bax =0
i.e. (a )
k
x b kl + (b )
m
(a x )ml = 0
ik im
Since b matrix is orthogonal

(a x )il + (b)ll + (b)ii (a x )il = 0
(a x )il { b ll + b ii } = 0
If b ll = b il then (a x )il 0 for i l otherwise (a)il =0 for i=l.
Since b matrix which is a diagonal matrix must be of even dimensions we
choose b matrix to be
1 0 0 0 0 0 - - -
0 1 0 0 0 0 - - -

- - - - - - - - -

- - - - - - - - -
b = - - - - - - - - -

0 0 0 0 - 1 -0 - -
0 - 0 - 0 - 1- - -

0 - 0 - -0 -1 - -
- - - - - - - - -

From the knowledge of Paulis spin matrices we find that there can be only
three 2 X 2 matrices anticommuting among themselves in pairs. It is impossible to
find a 4th 2 X 2 matrix that is anticommute with the Paulis spin matrices. But in the
case we need 3 a matrices and one b matrix to anticommute among themselves in
83
pairs. Therefore the minimum dimension of b matrix must be 6 X 6, 8 X 8, may be

possible for b matrix, but this will not give anything new and hence we choose the
minimum dimension the b matrix 4 X 4.
Therefore b matrix will be
1 0 0 0
0 1 0 0
b =
0 0 -1 0

0 0 0 - 1
In short we write
1 0
b= where each element is a sub matrix of order 2 X 2.
0 -1
Since (a x )il 0 for i l and (a x )il = 0 for i = l
We choose
0 ax1
ax =
a x 2 0
Further since a x 2 = 1
a x1a x 2 0
i.e. =1
0 a x 2a x1
ie a x1 a x2 = 1 and a x2 ax1 = 1
We choose both, a x1 and ax2 to be sx
0 0 0 1
0 s x 0 0 1 0
ax = =
s x 0 0 1 0 0

1 0 0 0
In a similar manner we choose

0 0 0 - i
0 s y 0 0 i 0
ay = =
s y 0 0 - i 0 0

i 0 0 0
0 0 1 0
0 s z 0 0 0 - 1
az = =
s z 0 1 0 0 0

0 - 1 0 0
84
It is to be mentioned that the above form for a and b is not unique. Any other
set can also be chosen to satisfy the above commutation relations.
However these two sets will be related by the singular transformation
a = s -1 a s
b = s -1 b s
where S is a non-singular free particle solution
9.5. Free Particle Solution
The Dirac Hamiltonian is

u u u r r
{
Ey = ac( + . b p) 2
y} mc
Since a and b are 4 X 4 matrices y must be a 4 X 1 matrix (or) It is a column vector

we choose
y 1 1 u
y u r r
yj = 2 =2 i(k. r w-
e t)
y 3 3 u

y 4 4 u
r r
i(k. r w- t)
or yj = j u e where j = 1,2,3,4
\ The Dirac equation becomes

r r u r u r r r
=
ih

{ uj }
i k ( -rw. t )
e { - ih a
c( . b+ ) mc }
2i(k.r
j u - w
e t)
t
r r u r r r
=
ih j u w(-i)
i ( k -. w
er )t
{ -ih a c( i )k b+ u }jmc
2 i ( k . r- wt )
e
hw uj = { c (a . hk ) + b mc }u 2
j
r r
Since hw = E h =
and k we
p have
urur
E uj =
{ c (a. p ) +b 2
mc} j u
u1 u1
u u
E 2
u 3
= {
c a x p x+ ay y p+ az + bp
z
2
} mc2
u3

u 4 u 4
85
u1 0 0 0 1 0 0 0 - i 0 0 1 0
u 1
0 0 0 0 0 i 0 0 0 0 - 1
E 2 = c
x + p
+ p + p
u 3 0 1 0 0 0 - i y 0 0 1 0
z
0 0
1
u 4 0 0 0 i 0 0 0 0 - 1 0 0
1 0 0 0 u1
0 1 0 u
0 2 2
mc
0 0 - 1 0 u3

0 0 0 - 1 u 4
Adding all the matrices
m 2 c c 0 p c( z p x) i -y p u 1

0 m (2 c c px +) i y p - cz pu 2
=
c p c ( px - iy p)- m c2 0 u 3

c( x +p y i )z p- c 0p - m cu2
4
E u1 = mc2 u1 + 0.u2 + cpz.u3 + cp u4

E u2 = 0.u1 + mc2 u2 + cp+ u3 - cpz u4
E u3 = cpzu1 + cp-u2 mc2 u3 + 0.u4
E u4 = cp+u1 cpzu2 + 0.u3 mc2 u4
[E- mc2 ] u1 0.u2 cpzu3 cp-u4 = 0 (A)

0.u1 + [E-mc2 ] u2 cp+ u3 + cpzu4 = 0 (B)
2
-cpzu1 cp-u2 + [E+mc ] u3 0.u4 = 0 (C)
-cp+u1 + cpzu2 - 0.u3 + [E+mc2 ] u4 = 0 (D)
Since the equation are homogeneous in uj the solution will exists only if the
determinant of the coefficient is zero.
E - m2 c 0- c p -z c p-
0 E - m 2 c- c + p c pz
=0
-c z p -c p- E+ m c2 0
-c + p cz p+ 0E m c2
By expanding the determinant we can show that
E2 = c2 p2 + m2 c4
2
E= cp+2 2 4
mc
86
Thus we find that both positive and negative energy values are possible for the
Dirac particle corresponding to E+ and E- we will have two solutions for both
2
Taking first E+ = + cp 2 + 2 4
, there are two linearly independent
mc
solutions for the positive energy.
Case (i)
u1 = 0 u2 = 1
c -p
From equation (C) u3 =
E + m 2c
-c zp
From equation (D) u4 =
E + m 2c
0
1
\ The wave function will be
c -p/ E +(m 2c)

-c z p/ +E ( m 2c)
Case (ii)
u1 = 1 u2 = 0
c zp
From equation (C) u3 =
E + m 2c
c +p
From equation (D) u4 =
E + m 2c
1
0
c z p/ E +(m 2c)
2
c +p/ E +(m c)
2
Similarly if we choose the negative root, E=-
- cp 2 + 2 4
, there are two
mc
linearly independent solutions for the negative energy.
Case (iii)
u3 = 0 u4 = 1
c -p
From equation (A) u1 =
E - m 2c
87
-c zp
From equation (B) u2 =
E - m 2c
c -p/ E -(m 2c)

-c z p/ -E ( m 2c)
0

1
Case (iv)
u3 = 1 u4 = 0
c zp
From equation (A) u1 =
E - m 2c
c +p
From equation (B) u2 =
E - m 2c
c z p/ E -(m 2c)

c +p/ E -(m 2c)
1

0
9.5.1. Normalization Condition

We use the normalization condition that u+ u = 1.
Case (i)
u+ u = 1
y+ y = (Nu) + Nu = N2 u+ u
0

c + p c - pz 1
N2 = 01 = 1
E + m 2c +E 2m c c -P/ E+m 2 c
2
-c zP/ +E m c
2 c 2 + pz 2p cz 2 p
N 1+ + = 1
( E+ m2 c )2+ ( E2 m 2 c )
2 c 2 xp 2+ y 2p c 2 zp2
N +1 + = 1
( E + m2 c )2+ ( E2 m 2 c )
c 2 p2
N 2 1+ 2= 2 1
( +
E m c )
Therefore the normalized wave function will be
88
0
1 r r
1 i(k. r w- t)
y1 = e
2
c p 2 c -p/ E +(m 2c)
1+
( E + m ) 2c 2 -c z p/ +E ( m 2c)
In a similar manner for the other three cases also the normalization constant
can be shown to equal to
1
0
1
y2 = 2
2 2
c p cp z /( E + mc )
1+ 2
( E + mc 2 ) 2 cp + /( E + mc )
c -p/ E -(m 2c)

1 -c z p/ -E ( m 2c)
y3 =
c 2 p2 0
1+
( E - m ) 2c 2 1
c z p/ E -(m 2c)
2
y4 =
1 c +p/ E -(m c)
c 2 p2 1
1+
( E - m ) 2c 2 0
E = c2p 2 + m 2c4
The energy correspond to positive as well as negative states

We find that y 1 and y 2 will correspond to positive energy solutions for spin
down and spin up particles
y1 = y +
y2 = y + -
In a similar manner the negative energy solutions y 3 and y4 will correspond
to spin down and spin up particles respectively.
y3 = y -
y4 = y - -
89
9.6. Meaning of Negative Energy States
It was found that necessary condition for the solution of Dirac equation is
E= 2cp 2+ 2 4
mc
The meaning of positive energy state is clear but that of negative energy is not
clear. It was at one time, suggested by Schrdinger himself as having no meaning and
hence to be excluded. However there were two fundamental objections for excluding
the negative energy states,
i) Mathematical and another
ii) Physical.
The exclusion of negative energy states will give rises to an incomplete set of
solutions and this is mathematically not valid. On the otherhand the Dirac equation
starts with positive energy states and there is a possibility of induced transition in to
the negative energy states. Therefore if the negative energy states are excluded there
will be contradiction, from Physicsts point of view. This has led Schrdinger into
insurmountable difficulties.
In order to overcome the above difficulties, Dirac postulated that
1. The positive energy levels form a continuum from +mc2 to + where as the
negative energy states form another continuum from mc2 to -.
2. In between mc2 and +mc2 there are no available energy levels.
3. Dirac proposed that all the negative energy states are completely filled up
normally.
4. The electrons in the negative energy states will not be affected by electric and
magnetic fields.
9.6.1. Consequences of Diracs Interpretation
The following consequences may arise due to Diracs interpretation.

1. The electrons in the positive energy states will not normally make transitions
to the negative energy states that are completely filled.
2. With the sea of electrons in the negative energy states, unobservable the
transition of one of these electrons to the positive energy states will manifest
90
itself, as a hole in the negative energy states. This hole is regarded as positions
and it behaves like an electron with the positive charge.
3. The fact that the negative energy states are completely filled up implies that
Paulis exclusion principle is obeyed. If the Paulis exclusion principle is not
valid, the negative energy states cannot be completely filled up. Since
otherwise many electrons rather than one electron will occupy level.
4. Feymann interpreted that negative energy states are electrons moving forward
in time (or) positions moving backward in time.
9.7. LET US SUM UP
In this lesson we had a discussion on the two prominent relativistic wave

equations namely K.G. equation and Dirac equation. Then we presented the free particle
solution of Diracs relativistic equation. Also we had a discussion on interpretation of
negative energy states.

Check your progress
1. If you were a Dirac How would you have predicted the existence of positron?
Explain
2. The K.G. equation has nothing to say about the spin of the particle Illustrate.
3. Explain how Dirac has interpreted the negative energy states.
4. Develop Diracs relativistic wave equation for a free particle and obtain its plain
wave solution.
5. Deduce relativistic Klein Gordon equation. Can we use this equation for pions and
kaons. Show that its probability expression is indefinite.
6. Establish the Diracs relativistic wave equation for an electron and calculate its
magnetic moment.
7. Obtain the eigenvectors for Diracs relativistic wave equation of a free particle.
8. Develop the relativistic wave equation for the particle of spin zero. Discuss the
difficulties involved in interpreting the probability density. Derive the expression
charge and current densities from the continuity equation.
91
9. Prove the following properties of Dirac matrices on the basis of Diracs theory
a. Square of matrices are unity
b. Eigenvlues of the matrices are 1.
c. Trace of the matrices are zero.
d. Dimension of the matrices are of even order.
e. The matrices anticommuting among themselves.
9.9. References
Publishing house.
92
LESSON 10
SPIN AND SPIN MAGNETIC MOMENT OF ELECTRON

This lesson deals with the expression for the spin of the electron and electron spin
magnetic moment.
10.2. Spin of Electron

We know that the commutation relation between the z component of orbital
angular momentum and the Dirac Hamiltonian is
urur
[ Lz , H] =y x , p -x y . ap + pb m
Here c is taken to be unity,
Now [ Lz , H]y = xx , -xp ya + yp ayp+
x z a
p z+p b m
= x y , a
p - x py .yp
x p x , ya
Here we have used the fact that x does not commute with px and y cannot
commute with py but they commute will all other quantities.
Also [ x, x p] = ,yy =p . ih
[ Lz , H] x= [x ,y p] p , yy
-x a px a py
= ih pxa
- x pxa y (1)
Here h is taken to be unity. But from Heisenbergs picture of time evolution,

dL z
we know that ih = [L z , H ]. (2)
dt
d zL
= p ya-x p xa . (3)
d t
y
This implies that Lz in not a constant of motion hence it is not conserved. In

other words, Lz is not the complete angular momentum and there must be another
part say Sz, such that (Lz + Sz) is conserved. We shall now verify that if Sz is taken
1 s z 0
to be where
2 z
= 0
z

s z
(4)

Then (Lz + Sz) commutes with H
93
Let us first note that,

0 s 0 sy
-ia x =ay i - ,
s x 0 s y 0
s xs y 0
= -i ,
0 s xs y
is 0
= -i z ,
0 is z
s 0
= z
0 sz
= z (5)
Now [S z , H ] = 1 [ z ,H ],
2
i
=-
2
[
a xa y , H , ]
i
=-
2
[
a xa y , a x p x + a y p y + a z p z + b m , ]
i
=
2
{
a-x ay, xa
p+ x a ,x ayp a
y , y }
i
=
2
{
(a-x ay ax -ax ax )yap+ xa( x a y-a y apy ax ay ) y },
i
=
2
{(
- a x2 -
ay )
ax 2- yap x a
+2 ( x a y+ a2px ay ) },
y
i
=- {- 2a y p x + 2a x p y },
2
= -i{- a y p x + a x p y },
= i {a
- x p y-a y p x }, (6)
From (1) and (6) we find that
[(Lz + S z ), H ] = 0 = i d (Lz + S z ).
dt
Therefore we find that the angular momentum is conserved only if we include the spin
angular momentum operator
1 1 s 0
Sz = = z (7)
2 z
2 0 s z
94
We see that J z = Lz + S z is a constant of motion and should be interpreted as the total

angular momentum of the system.
2 s z 0 s z 0
Since , = 0
z

s z 0
,
s z

1 0
= (8)
0 1
1
We find that the eigenvalue of = 1. Therefore the eigenvalue of S
z z = ( 1) are
2
1 1
+ and - .From these facts it is evident that the Dirac equation describes particles
2 2
1
of Spin such as electrons.
2
10.3. Spin Magnetic Moment of Electron (or) Dirac Equation in the

presence of Electromagnetic Field
A familiar way of introducing electromagnetic field in the Dirac equation is to

replace.
u r
u r r e A
p p - and
c
E E - ej
where A and j are vector and scalar potentials respectively. Therefore, the Dirac
equation becomes,
u r
u r r e A 2
(E - ej) y = ca - p b - + y mc
c
u r
u r r e A 2
ih - ej y = ca - p b - + y mc
t c
Assuming the vector function to be independent of time the wave function

can be written as
- i N( ,ER+ m 2c ) t
( )
y r, t = y (r ) e h
95
where E = EN.R + mc2 = Non relativistic energy + Relativistic energy
( )
y r, t = y ()r
- iN .E

h
e
R t

- i m2 c t
h
e
r - i m2 c t
= y' ( rt ) , e h
( )
wherey ' r, t = y (r ) e
-iE N.R t
h
is the non-relativistic wave form
y r - i m2 c - i m2 c
t
ih = ih yt ' y+'
(r,)
h
e
t t h
- imc 2
2
t
= ih + mc e h
y'
t
Substituting in the Dirac equation
u r

- imc 2
t u r u re A 2
- i m2 c
t
2 h
ih + mc - ej y '.e h = ca . p - b + mc y ' e
t c
Replacing the wave function y as a two component wave function y A and y B (large
&
small components)
u r u r
y A 0 s u 1r e A 0 y A
2
ih + mc - ej = c u r p. - + mc2 y
t y B s 0
c 0 - 1
B
u r
2 u r u r e A 2
ih + mc - ej y A = cs . p - y B +y A mc
t c
u r
y A u r u r e A
ih = cs . p -y f B + y eA (1)
t c
Similarly
u r
2 u r u r e A 2
ih + mc - ej y B = cs . p - y A y
- Bmc
t c
u r
2 u r u r e A
ih + 2mc - ej y B = cs . p -y A (2)
t c
Since
y A y B
ih = ih = ih {y ' (r, } t)
t t t
96
-iE N.R t

= ih y (r) e h

t
- iN .ER t
E
= ih -i N .R y h
e
h
-iE N.R t
h
= E N .R y e
Equation (2) becomes,
u r
u r u r e A
cs . p -y
{E N.R }
+ 2mc 2 - ej y B = c
A
u r
u r u r e A
cs . p y- A
yB = c
{ EN . R + 2mc2
f- } e
u r
y A u r u r e A 1
ih = cs . p - 2
t c +
EN . R2mc - f e
u r
u r u r e A
-cs . y-p + f Ae yA
c
Since EN.R <<< 2mc2 & assuming the magnetic field to be constant, ie j = 0,
EN.R + 2mc2 2mc2
u r u r
y A u r eu r A u 1r u e r A
ih = cs . p- s c 2 p y . - A
t c 2mc
c
u r u r
1 ue r uAr u er u rA
= s . p - s p . - (3)
2m c c
R.H.S can be reduced by the following relation
urururur u r u r u r u r u r
a .B a ( ) ( ).C = B . C+ a Bi . C
u r
where B & C commutes with a but not necessarily with each other. Here
u r
u r u r u re A
B = C=p - necessarily with each other. Now,
c
eA eA eA eA
s . p - s. p - = p - eA . p - eA + is p - p-
c c
c c c c
97
e u r u r u 2er u r 2
p2
c
(
- p .+ pA + A.A 2
c
)
= u r -u re u r u r u r
+ s i . +p . A
c
(p A. )
e2 2 - e
= p2 +
c 2
A
c
{ } - e
c

2 A. p - ih . A + is . (-ih A)

Equation (3) becomes
p2 2
e e urur
+ A2 A - p .
y A 2m 2 m 2
c m c
\ ih =
t i h ue r uh er u r u r u r
+ (
A . -s A .
2 m c 2 m c
)
Since A = B , the external magnetic field,
p2 2
e 2 e
urur
+ A A - p .
y A 2m 2 m 2
c m c
\ ih =
t i h ue r uh er u r u r
+ A. B-s .
2 m c 2 m c
This is nothing but, Schrdinger equation in N. R approximation.
The last term in the above expression gives the interaction of electron spin,
()
with the external applied Magnetic field B . Thus we find that the electrons behaves
eh
as a particle having a spin magnetic moment s where the spin of the e is
2mc
1
S= hs
2
Now it can be shown that
A =
1
2
( )
B r
Then
e
mc
A. p =
1 e
2 mc
( )
B r p
1 e
= B.r p
2 mc
e u r r r
B . L (where L = r ispthe orbital angular
= mc
momentum)
98
y A p2 2
e 2 e
u i r u r ehu r u ehr u r u r
\ ih = + A + - B . L A . Bs
- .
t 2m 2 m 2c 2 m c 2 m c 2 m c
p2 2
e i he u r eu r u r u r u r
= +
2m 2 m c 22
A2+ A. - B
m c 2 m c
.+ L h s ( )
1
Q S hs
2
y A p2 2
e i he u r ehu r u r u r u r
ih
t
= +
2m 2 m c 22
A2+ . -
A
m c 2 m c
B .+ L S 2( )
The last term gives the interaction of the orbital and spin angular momenta
with the external applied magnetic field B .
10.4. LET US SUM UP

This lesson deat with the expression for the spin of the electron and electron
spin magnetic moment

Check your progress
1. Obtain an expression for spin of the electron.
2. Derive an expression electron magnetic moment
10.6. References
6. Quantum Mechanics Chatwal Anand Himlayan Publishing house.
Publishing house.
99
LESSON 11
SCATTERING THEORY
11. 1 Aim and Objectives

In this lesson we are going to give an introduction to Scattering problems.
First we give some simple definitions for scattering amplitude, scattering cross
section. Also we will be presenting the kinematics of scattering process and the
relation between the scattering amplitude and scattering cross section.
11.2. INTRODUCTION
A beam of particles having a definite momentum when directed towards a

target (scattering centre) will undergo scattering in all possible directions. At very
large distance the scattered beam will be deflected normally outwards in the radial
direction. It is usual and convenient in practice to choose a co-ordinate system, in
such a way that the scattering centre as the origin and the direction of the incident
beam as Z-axis. The polar angles q and f of the scattered beam which describe the
scattering centre or scattering potential is chosen to be spherically symmetric. In other
words the spherically symmetric potential will only have q dependence and
independent of f.
11.3. SCATTERING THEORY
When a beam of particles of any kind is directed at matter the particles get
deflected from their original path since it collides with particles of matter. The
scattering experiments provide useful information about the nature forces, and
interaction between the particles of matter. In some collisions, the energy of the
incident particle does not change. Such collisions are called elastic collisions or
otherwise it is called inelastic collision.
100
11.4. KINEMATICS OF SCATTERING PROCESS
Consider a parallel beam of particles of a definite momentum is

directed towards a target. The particle gets scattered in different directions and finally
diverges. At larger distance from the target the particles are subjected to a radically
outward motion. Let us choose a co-ordinate system in which the origin is at the
position of the target and z-axis as the direction of incident beam. The polar angles q
Scattering Plane
Reference Plane
f
Target q
Incident beam
Co
ne
ofS
cat
teri
ng
of
an
gle
q
and f indicates the direction of scattered particle with z-axis as the polar axis. Here q
is the angle between the incident and scattered direction and is called as the angle of
scattering. The azimuthal angle f specifies the orientation of this scattering plane with
respect to some reference plane.
11.5. INCIDENT FLUX

Incident flux is defined as the number of particles crossing per unit area per
unit time, the area being normal to the incident beam. It is obtained by multiplying the
square of incident wave function with velocity.
| | P r2oyb a b+i l i t q
I= y d e n= s i t y V e l o c i t y
No.ofparticlesdistance
=
V o l u m e t i m e
No.ofparticles1
=
A r e a t i m e
101
11.6. DIFFERENTIAL SCATTERING CROSS SECTION

If DN is the number of particles scattered into a solid angle d W in time Dt,
then naturally DN will be proportional to the solid angle. The tune and the incident
flux. DN a W
d
a Dt
aI
ds
DN = d W t D I (1)
dW
ds
where the proportionality constant is called the differential scattering cross
dW
section. Therefore the differential scattering CROSS SECTION. is defined as the
Number. of particles scattered per unit solid angle, per unit time, per unit incident
flux. It has the dimension of area.
ds ND
\ =
dW dW t D I
The total scattering cross section is obtained by integrating the differential
cross section over all the polar angles q and f
p 2p
ds
s = q s qi nd fd
0 0
dW
For spherically symmetric potential which is independent of f, the total

scattering cross section becomes.
p
ds
s = q s i nqd
0
dW
11.7. RELATION BETWEEN DIFFERENTIAL SCATTERING

CROSS SECTION AND SCATTERING AMPLITUDE
For a spherically symmetric potential, which falls of rapidly than 1/r

asymptotic behaviour, the total wave function will consist of incident plane wave and
out going spherical wave.
r f (q e)i k r
u ( q,r =) ei k +z (1)
r
102
Where f(q) is called the scattering amplitude. The incident flux is given by
2
I i n =cUi n c J
2
= el k
z
J
= 1 J
= 1 J
The outgoing flux is given by

2
I =U
o u t o u t J
2
f (q ) k 2r
= 2
ei J
r
2
f (q )
= J
r2
2
f (q )
= Ii n. c
r2
\ the Number of particles scattered in the solid angle d, in time Dt is given by
DN = Outgoing flux x solid angle x Area x Time
2
f (q )
= 2 W I i n c d r2 Dt
r
2
= f (qW)I i n c d Dt (A)
But we find that the number of particles scattered in a solid angle d,
in time Dt is given by,
ds
DN = d W t D I (B)
dW
Comparing equations A & B.
ds 2
= f (q )
dW
Thus we find that the differential cross section is equal to the square of scattering
amplitude. Thus the scattering amplitude is related to the experimentally observable
ds
differential scattering cross section. Since has the dimension of area (m2 ), f() has
dW
the dimension of length (m).
103
11.8 LESSON END ACTIVITIES

Check your progress
1. What do you meant by incident flux?
2. Give the importance of scattering theory.
3. Define differential scattering cross section and total cross section. What is the
unit in which they are measured?
4. Give the schematic diagram of the scattering event.
5. What do you meant scattering cross section?
6. Obtain the relations between scattering cross section and scattering amplitude.
11.9. References
1.Quantum Mechanics A.K. Ghatak and S. Lokanathan McMillan

Publishers.
2.Introduction to Quantum Mechanics David J. Griffiths, Prentice Hall.
3.A text book of Quantum Mechanics P.M. Mathews and K. Venkatesan.
4.Basic Quantum Mechanics Ajoy. Ghatak Tata McGraw Hill Publishers
5.Introductory Quantum Mechanics Richard L. Liboff Addition - Willey.
6.Quantum Mechanics John L. Powell and Bernd Crasemann Narosa
Publishing house.
7.Modern Quantum Mechanics J.J. Sakurai.
8.http://thequantumexchange.org/
104
LESSON 12
SCATTERING THEORY
12.1 Aims and Objectives

In this lesson, we are going to study the various methods for analyzing
scattering problems namely partial wave analysis , phase shift analysis and Born
approximations. The above methods can be applied to low energy and high energy
scattering problem respectively.
12.2 . PARTIAL WAVE ANALYSIS
12.2.1. INTRODUCTION
The method of partial wave is elegant procedure for the analysis of elastic
scattering and low energy scattering. It is done in two steps. First wave function u(r,
) which represents the sum of incident and scattered wave is obtained in terms of
partial waves. Secondly the asymptotic value of this wave function is equated to the
wave function u(r, ) given by
r r
r f (q e)i
i k z
.k r
u ( q,r =) e + r (1)
r
A plane wave ei k zcan be expanded as a linear combination of spherical waves using
Rayleighs scattering formula.

ei =kiz l l (+j 2 1K )l (r )l P( c oq s ) (2)
l =0
Here jl ( K )r is a spherical Bessel function of order l. Pl (cos ) are the

legendre polynomials. Each term on the R.H.S. represents a spherical wave. The plane
wave is thus equivalent to a superposition of an infinite Number. of spherical waves &
the individual waves are called the partial waves. The waves with l =0, 1, 2,3 are
respectively called the s.wave, p-wave, d-wave and so on.
Substituting the value of jl ( K )r
105
s i k n-r p(l2 )
jl ( K )r=
k r
i ( K - rp 2) l -i (Kp 2- r )l
e e-
=
2i k r
Substituting in equation (2)
1z l i
ei =k i (l k
2+r e(1l -/ pi-)2 -ek) ( (pr - / cl 2P )o l s q)
2i k l =r0
This form shows that each partial wave can be represented as the sum of
incoming and outgoing spherical wave. Partial wave methods can be applied when the
potential is central and of finite range. This method is really useful whenever the
energy of the incident particle is low. Here the incident beam of particles a r e
represented by a plane wave with definite linear momentum with no definite angular
momentum. Hence a plane wave can be decomposed into infinite number. o f
components each of which corresponds to definite angular momentum (i.e) into
partial waves. The process by which a plane wave is decomposed into a partial wave
is called the partial wave analysis.
For a spherically symmetric potential having asymptotic behaviour, the total
wave function will consist of the incident plane wave and the outgoing spherical
wave.
f (q e)i k r
u ( q,r =) ei k +z (3)
r
Which is independent of . Using Rayleighs expansion for plane wave

ei =kiz l l (+j 2 1K )l (r )l P( c oq s ) (4)
l =0
Where jl ( K )ris spherical Bessel function and Pl ( c qo s is

) the legendre polynomial.
s i Kn -r( p/l 2 )
But jl ( K )r=
K r
Expressing in terms of exponential.
K-/ pr 2 l -) i (K
p - /r
ei ( e- 2l )
jl ( K )r= (3)
2i k r
il ( + lK2( -/ r-p 2 e-l1) (pi - / e2k) ) r f(l
i i
)ek q r
u ( q =, )r ( cq P o + sl ) (4)
l= o 2i k r r
The Schrdinger equation for the scattering problem can be written as
2 2m
+( E ) V-( u , rq=)
h 2
0

106
The Schrdinger equation can be separated into radial part and angular part.
The angular part of the solution will be spherical harmonics., Ylm (, ) which will
crumble to the legendre polynomial Pl ( c oq s for
) spherically symmetric potential
which is independent of .
The radial part of the solution Rl (r) will involve spherical Bessel function and
spherical Neumann function in the field free case. Therefore the total wave function
can be written as

u( q
,r )= R( )lr ( qlPc o s ) (5)
l =0
The solution of the radial Schrdinger wave equation

1 d r 2 d (l (R lr) 21 l ) +
2 +K
2
-(
U r) - (R )=r
l 0
r d r d r r
2m E 2 mV
Where K 2 = U2 , =(r )
h h2
The solution of the above equation (i.e) the radial wave function will be a linear
combination of jl ( K )rand hl ( K )r
- c o K-s r( p/l 2 )
Where hl ( K )r=
K r
However it should be noted that jl ( K )r is a well behaved function at r=0
where hl ( K )r is not so, since it becomes infinite at r=0. Therefore we choose the
radial wave function as R l ( r proportional
) to
Rl ( )ar c jlod lsK ( d l)-r ls hK
i n r( )
In this expression if we choose d l =0 for the field free case, then at origin, only
jl ( K )r will remain. At all other points both jl ( K )r and hl ( k rwill
) be present. This is
the reason why we associate c od ls with jl ( K )r and s i dnl with hl ( k r Rewriting

) the
above equation as
Rl ( )r= [ lAc od j s-ll (Kd ) slr l ih Kn ( r ) ]
Substituting the value of R l ( r in) equation (5) we get

U ( rq, =)R ql( oc s )
( rl ) P
l =0

= Al [ cj dol Ks -(l rd) slKi nlrh ( )Pl ] q(c o s )
l =0
107

s iK n- l[p r/ 2 ]K - c lpo r s ( / 2 )
= Al c do s l+P s d i l n l c( q o s )
l =0 K r K r
1
= Al [ c dolKs s -i nrpl( / l2+K) sd i lnr -c ol psP( / 2 )cq] ( o s )
l =0 K r
1
= Al . s Ki- n lp
r(l +/ P2 dl ) (c qo s )
K lr- 0
1 -Kl/p
= l
A i(
e -r+ 2l )di e-(lKl p(-/ rl P2d+o cs q) (6)
2i K l =r0
12.2.2. SCATTERING AMPLITUDE
Comparing the co-efficient of e -i k ,r in equations (4) & (6), we get

1 l 1
i ( l 2+P 1l =) qel ci p /
o P l s2 qcl ei p /o sel 2- - id l
A
2i l = 0 k r l2= 0i K r
The above equation is valid for all values of . Using orthogonal property of
lengendre polynomial, we can easily show that Al = l (i l+2 1e i)d Comparing the co- l
efficients of ei k rin equations (4) & (6), we get

1 l ( ) f q1
i ( l +2P l 1qle+ - i )p / e=lc 2 l-oei p/ sq Al 2 c id l
Po s
2i l = 0 k r r 2 l = 0i k r
Substituting the value of A l

1 ( ) f q 1 l
2i
i
l =0
l
( 2l+1i )P
k
l
+- p / 2
l c oqes
r r
= i+ ( l 2e 1 idlle) cq -eodli p / s2i Pcq l o
2i k l =r0
s
1 l 1 l p / 2 2 id l
i ( + 2l i
1l P) ql+c -e po/ 2 sqf ( ) = i +( l i2 e1l - )e c P lo sq
2i kl = 0 2i kl = 0
1 l l 1 l
f( q ) = i ( l +2e 1- i ep /) 2 2cidP-
l
lo sq i ( 2l+ 1i ) P l
l c oqes- p/ 2
2i kl = 0 2i kl = 0
1 l
= i ( l 2+ el1 - i)p /Pc2 l o- qes 2(id l 1 )
2i kl = 0
Substituting i l =i epl / ,2 -e2id 1ei=d eid - eid

l l l l
eidl -
= eid l [ e-
idl
eid-l]
= eid l . 2i s idnl
1 i p / l 2 i p / -l 2
f( q ) = e + ( l 2 1e ) cP lo seq 2idl si di nl
2i kl = 0
1
f( q ) = (+l2 1 P)l c oqe sldll s di n
k l =0
(7)
108
12.2.3. Scattering Cross Section in Terms Of Phase Shift
The differential scattering cross section is defined as

ds 2
= f (q )
dW
1 - ld
= +( l2 1 +l) '( 2P 1l ) Pcq o' se lcq oldel s sdl ' i' ln lsd i n
K2 l l '
Total Scattering Cross Section

p
ds
s = 2 ps iq n dq
O
dW
p
2p
s= +d( l2 1 +) 'l(d 2 1idl )e -idslq' i n esl i qln ' P
c lo s' qlcP o s s qdi n
k2 l'l o
2p 2
s= +( l2d 1 l)+ ( 2e ' id1lde )-idl s i n l s i dl n' ' l l
k2 l 'l 2l + 1
d l ' l= 1i lf= ' l

{by orthogonal property of legendre polynomial }
& d l 'l= 0 i l
f 'l

2p 2 2 2
s=
k2
+( 2l d1 ) s i n
l =0
l
2l + 1

4p 2
s=
k2
+( 2ld1 ) s i n
l =0
l (8)
This is the expression for total scattering Cross Section.
12.3. OPTICAL THEOREM

We know that the scattering amplitude is given by
1
f( q ) = (+l2 1 P)l c oqe sldll s di n
k l =0
A q = 0t c o s =l(P 1 =) 1 for
lq
P , all values of l.
1
f ( 0= )l( + 2 1e ) i dsl i dnl
k l=
Equating the imaginary part of the scattering amplitude

1
Im f (0) = (+2l 1 ) s i2dnl
k l=0
(1)
109
We know the total scattering cross section is given by

4p 2
s=
k2
+( 2l d1 ) s i n
l=0
l
4p 1
= . (+2 1l ) s i 2nd l (2)
k kl= 0
From (1) and (2)

4p
s= I mf ( 0 )
k
4p
Thus we find that the total scattering cross section is times the imaginary part of
k
scattering amplitude for q = 0. In other words the imaginary part of forward scattering
amplitude measures the intensity which the incident beam suffers because of
scattering.
12.4. PHASFSHIFT ANALYSIS
12.4.1 PHASESHIFT
The effect of introducing the potential into the picture is just to change the
phase angle of the normally outgoing scattered wave by an angle d l . We have just
seen that the radial wave function in the absence of potential is proportional to
s i Kn -r(lp/ 2 )
jl (K )r=
k r
Where as in the presence of potential it is proportional to
s i Kn -r(lp/ 2 )
jl (K )r= l + d
k r
Now in the expression for local wave number the radial Schrdingers wave equation
d2 u l ( +1 )
+ [k 2 ( u- 2 ) r ]+l (u =) r 0
d 2 r r
We find that for an attracting potential u(r) is negative and have k 2 + u(r) increases or
the wavelength decreases.
For an attractive potential the waves are pulled in and for a repulsive potential
u(r) is positive and the wavelength of the scattered wave increases and a repulsive
potential the waves are pushed out.
110
It is to be noted that a phase shift will be introduced only with in the potential of will
be maintained through out.
d l= 0
d l= +ve
d l= ve
RESULTS : -
(i) The green function for the operator

( 2 +k2) is given by

( 2 ) k2+ ( G -' r ) =rd( r-' r )
(ii) ( 2 k)2+ ei k =0z

-1 ei ( k - k0) r
(iii) G ( r -' r )=
4p r
12.5. BORN APPROXIMATION

This method is most useful for those problems whose potential are varying
rapidly. It is also used for analysing the high energy scattering problems. The
Schrdinger equation for the scattering problem is given by
2m
2 u ( r )+ E2( V( - r) ( = )r
U 0
h
111
2 m E 22m
letting = k 2, V ( =r ) U( r)
h2 h
2 u ( r +)2 k (u =)r U
( )r u( r)
( 2 ) k2+( u ) r= (U )r (u r) (1)
Noting that ( 2 +k2) e lk 2 is zero, we can write the solution for Schrdinger equation as

U(r) = ei k G+z r ( -r ' U) (r ' u) r( d' )r ' (2)
This can be easily verified by operating ( 2 +k2) on both the sides of equation (2)

( 2) k(2+ )u =0r (r +' rd ) -U( r' )u (r ' d ) r
( 2 ) k2+( u ) r= (U )r (u r)
Thus we find that the expression for u(r) given by equation (2) is valid.
Now denoting uo ( r =) ei k ,z we have

u(r) = uo (r )+G( r '-r) U( r' )u (r 'd ) r ' (3)
Here we find that right hand side contains u ( r' (i.e.)

) a function of u which we are
seeking to determine. Now by replacing r by r ' , r ' by r ' , we
' get.

u ( r' =) u(o r' +) G( 'r '- r' ) U( 'r ' u) (r ' d' ) r ' ' (4)
Substituting (4) in equation (2)

( ' r )u ( r ' d ) r +' G( r' -r) (U ' r) G
u ( r )= uo( r )+ G( r' - r)o U (- ' r ' r' )U ( r' u' )r ( d' r' )d ' r' '
The above series can be extended further by replacing r by r by r ' b' y' r' &' so on.

u ( 'r ' =)u (o 'r ' +)G( ' r' -' r' ' U
) ( 'r ' 'u) (r ' 'd' ) r' ' '

) uo( r )+ G( r' - r)o U
u( r = ( 'r )u ( r ' d ) r +' G( r' -)r ( U' )r (-G' ' r' )r ( U' ' r) u( ' r ' d) r' d' r'

+ G( r'- )r ( U' ) r-( G' ' r' ' r' u)( r' ' G) ( r' -'r ' U
' ' )r ( u' ' r' )d( r' ' d' )r' 'd' 'r ' '
Such a series is called Born series or Neumanns series. This expression is evidently a
perturbation expansion of U(r) with the potential treated as perturbation. If the
incident energy is very large, then the series will be converging rapidly. In such a
case, only the first few terms will contribute and it will be sufficient to terminate the
series after two or three terms. Thus for first Born approximation, we will have two
terms.
112
Therefore in the first Born approximation the wave function will be.

u ( r )= u0( r)+ G( r' - )r0 U
( ' r )u ( r ' d ) r '
Substituting for the Greens function and for uo ( r )we have,

i k z -1 ei k rei ok r
u( =r) (e )U ( -r ' u) (r 0 'd ) r '
4p r

i k z 1 ei ( k - ko) r
=e U - r u - ( r ' ) d ( ' r ) '
4p r
Comparing the above expression with total wave function

f B (q e)i k r
U ( qr, =) ei k +z
r
-1 - l ok r
f B (q e ,f=U) ( r ' u ) r( d' ) r '
4p
Dropping the primes, we get
-1
f B (q e ,f =U) r( - i ok)u r (r ' d ) r '
4p
Assuming u(r) = ei k ,r the above equation becomes.

i
-1 - i ( Ko- K
)i r
f B (q , f=) e
4p

Defining K o - Kl = K the momentum transfer function.
-1
f B (q e f,= u ) -ri (k. rd ) r (7)

4p
The directions of K o - K l are shown below
Ko
K = ko -ki
ki
From equation (7), it is clear that for the first amplitude depends only on the
momentum transfer and not on the initial and final momenta and the angle of
scattering. For a spherically symmetric potential, the potential will be independent of
angle f and therefore.
p 2p
-1
f B (q e , = U)f r( - i)rqk r
sd 2
qi r nd f d
4p 0 0 0
113
p
-1
= e - i kUr ( r ) 2r s di rqn dq
2 0 0
p
-1 2 ( c o i k -sr q)
= r U( r) -d ( r c e od s q)
2 0 0
p
-1 ( c- eoi k r s q )
= r 2 U( r )d r
2 0 i k r 0

-1 2 - ei k r ei- k r
= r U( r )d r
2 0

i k r

-1 ei k -r ei- k r
f B (qr =
) U ( r ) ds ri kn r since s i kn =r
k 0 2i
This is the Borrn amplitude for the spherically symmetric potential.
12.5.1 Validity Condition For Born Approximation

Born approximation is valid only when the total wave function is greatly
different from the incident wave function.
Let g ( r<) <e i k z
g ( r)2 <| <1
Where g(r) is given by

-1 ei (k -r' r)
g ( =r ) U ( r ' e ) di k z
r
4p r -' r
Substituting the value of U(r), we get

-1 ei (k -r2' r) m i k z
g ( =r ) V ( r ' 2 e ) d r
4p r -' r h
-1 ei (k -r' r) m
V ( r ' ei k' )dz r'
2p r -' r h 2
=
In several cases g (r a2)<t t <1h e o r g i n

m e i k' r
g ( = V0 - )2r (e ' d i) k' zr '
2p h r'
Dropping the primes.

-m e i k r i k z
g ( =V0 2r) (e )d r
2p h r
2
2 m ei k r i k z
g( V= 0 r2 ) <e ( d )<r 1
2p h r
This is the validity condition for Born approximation.

114
If V ( r )is spherically symmetric, we have,

1 2p 2
2 m ei k .r ei kwr
g( = 0 r2
V ) r (dr r ) d2 wd f
2p h 0 -1 0
1 2
2 m 2 i

k r i k rw
g( e=0 r2) 4 Vp( r2 d) r(w
e) d.
4p h 0 1 -
2
1
m2 i k r

i k r i k rw

=
h4 e e V r( )r . d r e wd
0 0 1 -
1 2

i k w
r
r e

m2
i k r

= 4
h e V (r )r . d

0
i k r
-1

2
ei -e
k r i- k r
2
m
i k r
= 4 e V (r )r . d r
h 0 i k r
2
m2
= 2 4
e 2i k)er 0 V
(- . (r d) r
k h 0
2
2 m2
g( 0e= 2) 4V( - 2i1rk r ) <d . (r < ) 1
k h 0
according to Born approxination.

12.6 LET US SUM UP
From this lesson one can get a clear idea on partial wave analysis , phase shift
analysis and Born approximation. Also the optical theorem was presented. Once this
theory are understand clearly one can extend this methods to scattering problems in
atomic and nuclear physics.

Check your progress
1. What is called a partial wave ? Why it is called so.
2. Define the term phase shift.
3. What is called Born series.
4. Write a brief note on optical theorem.
5. Enumerate the two conditions for the validity of born approximation.
115
6. Obtain an expression for the Born approximation amplitude and hence

derive the Rutherford formula for Coulomb scattering.
7. What do you meant by partial wave analysis? Employ this method to
derive an expression for the scattering amplitude in terms of phase shift.
8. What are called Green function? How they are used in scattering theory?
9. Define phase shift and hence describe partial wave analysis in scattering
theory.
12.8 References
1. Quantum mechanics K. Sankar Narosa publishing house
2. Quantum Mechanics A.K. Ghatak and S. Lokanathan McMillan
Publishers.
3. Introduction to Quantum Mechanics David J. Griffiths, Prentice Hall.
5. Basic Quantum Mechanics Ajoy. Ghatak Tata McGraw Hill Publishers
6. Introductory Quantum Mechanics Richard L. Liboff Addition - Willey.
Publishing house.
8. Modern Quantum Mechanics J.J. Sakurai.
9. http://thequantumexchange.org/
116
LESSON 13
APPLICATIONS OF SCATTERING THEORY
In this lesson, we are going to discuss in the scattering by different potentials

namely Coulomb , Yukawa potential. Also we will be concentrating on central field
approximation methods, Thomas Fermi Statistical Model and Hartrees Self
Consistent Fields.
13.2. Scattering By Coulomb Potential

The potential given as
ze 2
V (r ) = - (1)
r
The Born Amplitude is given by

1
f (q ) = - r U (r ) sin kr dr
k 0

2 m 2 sin kr
= ze dr
h2k 0
r
2 mze 2 1
= .
h2k k
2mze 2
= 2 2 . (2)
h k
q
We know that momentum transfer k = 2k sin
2
2 mze 2 mze 2
\ f (q ) = = (3)
2 q
2 2 2 2 q 2
h 4k sin 2h k sin
2 2
Differential scattering cross section is given by

117
ds 2 m 2 z 2e4
= f k (W) =
dW q
4h 4 k 4 sin 4
2
m 2 z 2e4
, where p = hk
= (4)
q 4 4
4 p sin
2
This coincides with the classical Rutherford formula for coulomb scattering.
13.3. Yu kawa Potential

The potential of the form
e -a r
V ( r =) Vo
r
where Vo and a are constants is called yukawa potential. It is an example of a central

potential. Here the potential depends only r and not on q , f. It was orginally
introduced to represent the interaction between fundamental particles.
The Born amplitude is defined as
-1
f B (q e f,= U ) -ri k( r d ) r
4p
-1 m2 - i k r
4p h 2
= e V r( d) r
-1 mVo - i k re -a .r
2p h 2
= e d r
r
a p2 p
-1 mVo - i k cr q o e -as.r 2
=
2p h 2 e
0 0 0
r d
r
r s idqn qd f
a p
-1 mVo
= 2p e -(a rr i-
dc
k q-r ( c o s )
r- qo e ds )
2p h 2 0 0
m vo -a r e -i k(-rc oq s )
= e r d r -i k r
h2 0 0
-i k r
mV ei k -r e
= 2o e -a dr r
h 0 i k r

-2mVo -a r
=
h 2 k e
0
ks ir n d r
Evaluating the integral and simplifying, the scattering amplitude becomes

2mV 1
= f B (q= , f )- 2 0 2
h a + 42 k s q2i n / 2
118

Here q is the angle between the incident wave vector k and the unit vector r in the
scattered direction. From the above expression, it is clear that, the scattering
amplitude is independent of angle f . This independence is a general result for elastic
scattering by a central potential.
The differential scattering cross section can be written as
ds 2
= f (q ,f
dW
4m 2V 2 o 1
=
h 4 (a 2
4+ s2 k i q n2 / 22 )
Thus , we can see that, the differential scattering cross section varies with k and so it
also varies with energy E. In the limit of zero energy (k 0), the differential cross
section is
4m 2V 2 o
s=
h 4a 4
Thus the differential cross section is independent of q as well as of f . Except at zero
energy, s peaks in the forward direction ( q =0) and decreases monotonically as q
varies from 0 to p , which is shown is fig. below. It is to be noted that, with in the
Born Approx, the differential cross section is independent of sign of Vo and gives the
same result if the yukawa potential is attractive (V0 <0 ) or repulsive (V0 >0 )
1.0
0.9
(s / )
4m2 v 2 ao4 h4
0.1
0.8
0.2
0.7
0 p
q
119
13.4. Application to Atomic Structure : Central Field Approximation
The starting point of calculations on all except the lightest atoms is the central
field approximation. This assumes that each of the atomic electrons moves in a
spherically, symmetric potential energy V(r) that is produced by the nuclear and all
the other electrons. The approximation is a good one if the deviation from the V(r) for
one election produced by close passage of other electrons is relatively small. This is
actually the case, since the constant nuclear potential is of the order of Z times as
large as the fluctuating potential due to each nearby electron, and the latter varies
quite slowly (inversely) with the separation distance. The two principal problems are
the correction of the approximate results obtained from it. Before considering these
problems, we discuss some general properties of the central field.
The potential energy V(r) for a neutral atom has the coulomb form c2 /r at a
great distance r from the nuclear, since the removal of the electron whose potential is
being measured leaves a singly charged positive ion. The electron in the hydrogen
atom, for which the potential energy is c2 /r at all r has an infinite number of bound
energy levels characterized by the quantum numbers n, l and m. An infinite number of
energy levels is also expected for V(r), since for large n the electron wave function is
small near the nucleus and only the form of V(r) of large r is significant .An important
difference between the two situations is the degeneracy between states of the same n
and different L that occurs in hydrogen is removed in a noncoulomb central field. This
is because the electrons that have smaller angular momentum penetrate closer to the
nucleus, and V(r) is stronger (more negative) than c2 /r there, since the nucleus is less
completely screened by the other electrons. Thus for given n, the states of lowest l
have the lowest energy. The degeneracy with respect to m is not affected, since this
occurs whenever the potential is symmetric.
Because of the spin, four quantum numbers n, l , me and ms are required to
specify the state of an electron in a central field. The orbital quantum numbers l and
1
ml are the same as l and m in the hydrogen atom, ms = specifies the spin
2
orientation and n is a natural generalization of the total quantum number that appears
in hydrogen.
120
The equation l = n = n+ l +1 shows that n - l -1 is the number of nodes of the

radial part of the hydrogen wave function; this definition of n is carried over to the
general central field, so that l does not exceed n-1.
13.5. THOMAS FERMI STATISTICAL MODEL
Two methods have been used for the determination of the potential energy
V(r). The first of these Thomas. Fermi statistical model assumes that V(r) varies
slowly enough in an electron wavelength so that many electrons can be localized
within a volume over which the potential charges by a small fraction of itself. The
electrons can then be treated by statistical mechanics and obey the Femi Dirac
statistics. At normal temperatures the thermal energy kT is very small in comparison
with V(r) every where except at the edge of atom where the chance of finding an
electron is small. In this case, the Fermi Dirac statistics requires that the electron
states fill in order of increasing energy, as assumed.
The number of electron states in a cube if edge length L at the walls of which
the wave functions obey periodic boundary conditions was computed to be
3
L
dk x dk y dk 2 . This must be multiplied by 2 to take account of the two
2p
possible spin states; then the number of states for which the momentum P = hk
is less than or equal to P0 .
p0
3 h p 2p 3
L 2 Po L3
2 d 0 k dk sin q dq df =
2p 0 0 3p 2 h 3
If all these states are occupied, the number of electrons per unit volume whose
2 3
Po Po
kinetic energy does not exceed is
2m 3p 2 h 3
Now the maximum kinetic energy at any distance r form the nucleus is V(r), since
otherwise the electrons would escape the atom. We thus obtain a relation between the
volume density of electrons, n(r) and the pontentialenergy :
n(r) =
[- 2m v(r )]3 2 (1)
3p 2 h 3
121
The electrostatic potential V(r)/e is also determined by poissons equation. in terms

of the charge density en(r) :
1 2 - 1 d 2 dv
- v= r = 4p en (r ) (2)
e e r 2 dr dr
Equations (1) and (2) and two simultaneous equations for n and r, the boundary
condition of the solutions can be expressed in terms of v along for a neutral atom of
atomic number z. as r0, the leading term in the potential energy must be due to the
nucleus, so that V(r) - Ze2 /r. As r there must be no net charge inside the
1
sphere for radius r, so that v falls off more rapidly than and rv (r) 0. The
r
boundary condition at is different from that assumed earlier in this section, where v
was taken to have the asymptotic form c2 /r. The v discussed earlier is the potential
experienced by one of the atomic electrons, where as the Thomas Fermi potential is
that experienced by an infinitesimal test charge. The diff. between the two potentials
emphasizes the statistical nature of the approximation made Thomas and Fermi. The
soln. for v is exact in the limit in which m becomes infinite and c becomes zero in
such a way that m3 e4 remains constant then the electron wavelength becomes zero and
the density of particle becomes infinite. In this limit the potential is constant over
many wavelengths and enough particles are present so that statistical mechanics can
be applied
13.5.1. EVALUATION OF THE POTENTIAL
Elimination of n(r) from (1) and (5) leads to an equation. for V(r).
3
1 d ( d2 -V) 4e 2 [ -2m V
( r)] 2
r = (3)
r 2 d r d r 3
3p h
we put
- z 2e
V(r) = c r =b x
r
2
1 3p 3
h2 0 . 8 8ao5
b= = 1 1
(4)
2 4 m 2
e3 z 3
z
where a0 = h 2 with the substitutions equation (3) becomes

me 2
122
1
2
d 2c 3
2
x =c
dx 2
c = 1 at x = 0 c = 0 at x = (5)
The most accurate solution of Equation (5) was computed by Bush and
Caldwell with the help of the original differential analyzer and is expressed form of
numerical table.
Equation (4) show that the radius of an atom is inversely proportional to
the cube root of atomic number, if this radius is interpreted to be that of a sphere that
enclose a fixed fraction of all the electrons. These equations can also be used to show
that the Thomas Fermi approximate improves with increasing z. The potential at the
atomic radius is proportional to z 1/3 so that a typical electron wavelength is
proportional to z 2/3 . The distance over which the potential changes by a definite
fraction of itself is proportional to the atomic radices or z -1/3. Thus the fraction change
of the potential in an electron wavelength is proportional to z -1/3 and decreases with z.
Moreover since the number of electrons is equal to z, the use of the statistical method
is better justified s z increases.
13.6. HARTREES SELF CONSISTENT FIELDS

The second method for obtaining a central field is due to Hartree. This
model assumes that each electron moves in a central field that can be calculated from
the nuclear potential and the wave functions of all the other electrons, by assuming
that the charge density associated with an electron is e times its position probability
density. The schrodinger equation is solved for each electron in its own central field,
and the resulting wave functions made consistent with the fields from which they are
calculated. Thus the kth electron is described by a normalized wave function uk (rk )
that is a solution of the equation
- h2 2 ze 2 2 e2 3
k - + | u i (r j ) | d r j u k (rk ) = k uk (rk ) (1)
2m rk jk rj k
where rjk = |rj-rk |. If there are z electrons in the atom, equation (1) constitutes a set of z
simultaneous nonlinear integral differential equations for the z functions uk (rk ). It is
therefore not feasible to solve these equations. directly, and Hartree used a method of
successive approximation.
123
A potential energy that approximately represents the second and third

terms in (1) is assumed, electron wave fns. Computed and new potentials for each
electron found from these wave functions. This process is continued until the
potentials are self consistent to a high order of accuracy. The principal
approximation made is the averaging of the potential energy given as the third term in
(1) over the angles of rk to make it spherically symmetric. The solutions of (1) can
then be expressed as products of radial functions and spherical harmonies. A further
simplification is made so that the 2 (2l +1) or fewer electrons in a shell all move in the
same potential and have the same radial wave function.
It is apparent that the Hartree approximation neglects correlations
between the positions of the electrons, since the entire wave function for all the
electrons is assumed to be a simple product of one electron functions.
y (r1 , r2 , ........, r2 ) = u1 (r1 ) u2 (r2 ) ................. u2 (r2 ) (2)
It is also clear from (2) that antisymmetrized wave functions are not employed. The
antisymmetry is considered only in so far as the Quantum numbers of the one
electron states uk are chosen in agreement with the exclusion principle.
13.6.1. CONNECTION WITH THE VARIATION METHOD
We now show that the Hartree approximation results from an optimum

variation calculation with the trial function (2). The wave equation with inclusion of
interelectronic interactions but neglect of spin orbit terms is
Hy = Ey
h 2 2e 2 e2
H = k
- + r> (3)
k 2m j kr k j k
where j > k implies a double summation over all different pairs of indices j and k we
wish to minimize the expectation value of H,
From (2) and (3) we obtain
3 3
. . . . . . y. . . H*y. . . .d. . . . r. . . . d. . . . .r.
1 2
* -h 2 2 z e
= u k ( rk ) -u2k (kr
3
d)
k k r
k 2m r k
e2
+ > u *j( r k )u k (r* ku) jr( 3 uj ) rk ( dk3) r j d k r (4)
j k rj k
124
since the uk are normalized. The optimum y is obtained by varying each of the uk
separately to minimize equation (4). The only dependence of equation (4) on a
particular one electron function uk is through the terms.
* -h 2 2 2 z e 3
uk ( rk ) uk -(kr d)
k r
2m r k
e2
+ u *j( r j )u k (r* ) j r( 3 ju) rk ( dk 3) r j d k r
ku
j k rj k
= u k* ( rk ) H k u k ( rk ) d 3 rk (5)
-h 2 2 z 2 2 e 2
e
Hk k - +j | u k( )r3 | dj r
2m rk j kr j k
The integral in equation (5) is the expectation value of the operation Hk for the
function uk . This is a minimum when uk is an eigenfunction of Hk that corresponds to
its lowest eigenvalue k .
Hk uk = k uk (6)
Since equations (1) and (6) are identical, we see that the Hartree wave
functions are the best from the point of view of the variation method that can be
written in the form of equation (2)
The energy associated with this wave function is just the integral in equation
(4) which can be written with the help of equation (1)
3
..................... u * H y d r1 ................... d 3 r2 = k
k
e2
- > | u j (r j ) | 2 | u k (rk ) | 2 d 3 r j d 3 rk
j k r jk
The electrostatic interaction terms between electrons are counted twice in the
summation over k and so have to be subtracted to give (7). Thus the energy of the
atom is not just the sum of the k , although each k . is roughly the energy of removal
of the kth electron. This last is not strictly true, since the removal of an electron alters
the self consistent fields and hence the wave fr. s for the remaining electrons.
However, k is found to be an especially good approximation to the energy of
removal in the case of an inner electron (x ray level).
125
13.7. LET US SUM UP

In this unit we have discuss with the coulomb and the yukawa potentials. Also
we have studied Thomas Fermi Statistical Model, Central Field Approximation,
Hartrees Self Consistent Fields.

Check your progress
1. What is the range of Coulumb potential?
2. Define yukawa potential.
3. Write down Hartree Fock equation.
4. Discuss in detail the problem of Yukawa potential.
5. Discuss in detail the problem of Coulommb potential. Obtain an
expression for scattering cross section.
6. Explain the central field approximation method in detail.
7. Explain in detail Thomas Fermi Statistical model.
8. Discuss Hartree Fock self consistent model.
13.9 References
2. Publishers.
5. Tata McGraw Hill Publishers.
9. Publishing house.
126
LESSON 14
APPLICATION OF SCATTERING THEORY TO
ATOMIC AND MOLECULAR STRUCTURES
14.1 AIMS AND OBJECTIVES

In this lesson we are going to study the application scattering theory to atomic
and molecular structure physics.
14.2ALKALI ATOMS
The ground state configuration of an alkali atom consists of a series of full

shells followed by a single s electron and so is 2S . The inner rare gas
configuration is so stable that all but quite high excited states of the atom involve only
the valence electron. Thus the alkalis can be treated to quite good approximation in
terms of a model in which a single electron moves in a spherically symmetric
noncoulomb potential energy V(r).
14.2.1 DOUBLET SEPARARTION
The configuration of an alkali atom can be specified by a single pair of

quantum numbers nl. Since there is only one electron, the perturbing electrostatic term
ii does not appear. In the absence of external fields the hamiltonian, including the spin
orbit energy is
-h 2 2
urur
H= V +
r +
(r ) L( ) S. (1)
2m
where
1 1 dV
(r ) = 2 2
2m c r dr
The total angular momentum J = L+S of the valence electron is a constant of
the motion, so that states can be designated by jm instead of ml ms, where J2 =
j(j+1)h2 and Jz = mh .The states of different j have different energies, but there is
still a (2j +1) fold degeneracy due to m. The removal of the m degeneracy by a
magnetic field is discussed later.
127
The difference in energy between states of different j is due to the L.S term in
(1) and can be found from its expectation value or diagonal matrix element.
2 u r r
J 2 = LS
( + 2 ) L= 2 + S + 2 L.S (2)
since e, j and s are all good quantum number (s = ). Equation (2) can be solved for
the diagonal matrix element of L.S.
u r r 3
lL.Sl
j j = [j +(j1) +l (l1) -J 2k (3)
4
Now if L is different from o,j can be either L + or L . Thus the first

order perturbation arising from (r) L.S. is
1 l nl 1
if j = l +
2 2 (4)
-1 1
(l + 1) nl if j = l -
2 2
nl = h 2 |R nl (r ) | 2 (r ) r 2 dr l >0
0
where Rnl (r) is the normalized radial part of the unperturbed eigenfunction associated
with the nl configuration. Since V(r) represents an attractive potential energy, (r) is
positive and nl is positive. Thus equation (4) shows that the state with higher j has
the higher energy. The pair of states is called a doublet the doublet structure
characterizes all the moderately excited levels of the alkali atoms except those for
which l = 0 in which case j can only be .
The doublet separations can be calculated from equation (4) if radial function
is known using the generating function for the associated lauguerre polynomials.
lj

h2 z 2 e 1 2
nl = R n r(l d) r
2 m2 2 c 0
r
128
e2h 2 Z 4
= (5)
1
2 m 2 c 2 a o 3 n 3 l l + (L + 1)
2
This is valid only for l >0 ; the singularity in (r) at r 0 makes the integral for no
diverge there, so that the perturbation approximation is not valid. It follows from (4)
and (5) that the doublet separation is proportional to n-3 and this in fair agreement with
observation. The absolute value of the doublet separation and its dependence on l are
not given at all by this simple theory, since the effective z is difficult to estimate and
depends marked by on l because of penetration.
14.2.2 . DOUBLET INTENSITY
We now calculate the intensities of the two lines of the allowed doublet 2P
3 1 1 1
2S & 2P 2S under the assumption that the radial wave functions
2 2 2 2
are the same for the two excited 2P states. Transitions of this type give rise to the
principal series in the alkali spectra. From the equation.
2
4 e 2 k 2 wkn 2 4 e 2 wkn
| < k | r | n >| = | < k | r | n >| 2
3h c 3h c3
the spontaneous transition probabilities and hence the observed intensities if the two p
states are equally likely to be occupied, are proportional to the squares of the dipole
matrix elements
The dependence of the two excited 2p states and the ground 2S state as the
angular and spin co-ordinates of the electron is obtained by finding linear combination
of products of the four spherical harmonics y1 , 1(q, f), y1 , 0 (q, f), y1,-1 (q, f), y0 ,0 (q, f)
and the two spin wave functions (+) and (-) that are eigenfunctions of J2 and Jz. These
combinations can be obtained from the Clebsch Gordan co-efficient.
129
3
m = y+ ( ) 1 1,
2
-1
1 1
= + 3 2 +22 ( y1)-, 0( y1 ), 1
2
-1
-1 1
2 p3 = 3 1 2-, y (2 0 +2 ) 1 y( +, )1
2
2
-3
= - y1 , 1
2
1-1
1 22

m = - y 3 + ( )+ 1 y2, 0-( 1 ), 1
2
-1 1
-1 2
2

2 p 1 =3 1 ( -,y ) 0 -2 1 y( +, ) 1 -
2
2
1
2s1 = 0(m , +)y 0
2
2
-1
= - ( )y0 , 0
2
The wave functions equation (6) can be used to calculate the matrix elements of
x = r sinq cosf y = r sinq sinf and z = r cos q. We assume that the radial functions
associated with equation (6) are all the Same, so that the radial part of the matrix
element integral is a common factor throughout. The angle parts of the integrals are
easily evaluated by making use of the explicit expressions for the ys in terms of q and
f. The products of spin fns, follows the simple rules (+) (+) = 1, (-) (+) = 0 etc., In this
way we obtain the following values for the squares of the magnitudes of the indicated
1
matrix elements, expressed in units of of the common radial factor :
18
3 1
m = t =o m = | x 2| =| y2| 3 z2=| | 0
2 2
3 1
= t= o - x =| 2|y |=2 | z 2| = | 0
2 2
1 1 2
2 p 3 2 1 s | =t | o| = | x 0 = y 2| |2
z= 4
2 2
2 2
1 -1
= t =o | x 2|= | y 2| 1 =z2 | | 0
2 2
130
1 1
m = to m = | x |2 = | y |2 = 0 | z |2 = 2
2 2
1 -1
= to | x |2 = | y |2 = 2 | z |2 = 0
2 2
-1 -3
Similar results are obtained for the transitions that start form m = and .
2 2
It follows from equation (7) that the sum of the intensities of all the lines that
originate on each of the four 2P3/2 states is equal to 6, in the above units. It is
expected that these sums are equal since the four values of m differ only in the
orientation of the angular momentum, and this should not affect the intensity.
However the total intensity form each of the two 2P1/2 states is also equal to 6. The
equality of total intensities from each state formed form a given L and S is a general
property of LS coupling; this makes the absolved intensity, which is that from all the
states that are degenerate with respect to m, proportional to 2J+1. In the example
considered here, the two lines for the doublet have intensities in the ratio 2:1. This is
observed for the lowest doublets of the alkalis, although for the higher doublets the
intensity ratio exceeds 2. This is because the spin orbit energy actually mixes
different configurations (2p states with same j but different n ) ; the amount of mixing
is different for the two j values, so that the two radial functions are not the same. A
small admixture of the low- intensity upper states in the high intensity lowest 2p
states has little effect, whereas in the opposite case there is a large effect on the
doublet intensity ratio.
14.3. Application to Molecular Structure: Hydrogen Molecule
It is clear that two distinct problems arise in connection with molecular

structure. The first is the solution of Schrdinger equation to obtain electronic wave
functions and a potential energy function of the nuclear coordinates. The second is the
solution of nuclear motion. The first problem can be solved only in the simplest cases.
As an example, we now consider in outline an approximate solution for the hydrogen
molecule due to Hietler and London.
131
14.3.1 .Heitler and London Method
The only nuclear coordinate Rj that appears in the equation

h 2 n 2
-
2 m
i V+R u j (i r= ) jU (j RR ) u i( r ) (1)
i =1
in the case of the hydrogen molecule in the magnitude R of the distance between the
two hydrogen nuclei. The Hamiltonian is that given in equation
H = H0 + H '
h2 2
e2 e
H 0 = -1 ( 2 2+ 2) - - (2)
2m r 1r 2
e 2 e2 e2 2
e
H'= + - -
R 1r 2 rB1 r2A
however, R is no longer large in comparison with a0 = hm2 / 2e so that the

approximations implied in equation (2) are no longer useful. Nevertheless, an
approximate wave function based on a simple product of two ground state hydrogen
atom functions gives remarkably good results. The reason for this is that exchange
degeneracy is taken into account; the degenerate wave functions for which electron 1
is on nucleus A and electron 2 on nucleus B, and for which electron 1 is on nucleus B
and electron 2 on nucleus A are both used at once. The new feature of the work of
Heitler and London was the recognition that an appropriate linear combination of
unperturbed degenerate wave functions gives a significantly lower energy than the
separate wave functions; it is the basis of the present day theory of homopolar
binding in molecules. This property of degeneracy is sometimes referred to as
resonance. An analogous situation is that in which an interaction between two
classical oscillators that are in resonance (same unperturbed frequency) gives rise to a
normal mode that has a lower frequency (and also one that has a higher frequency). In
a similar way, an interaction between two resonant (degenerate) state in quantum
mechanics gives rise to a lower energy eigenvalue (as well as to a higher one). There
may of course be more than two degenerate unperturbed states, and the degeneracy
need not be of the exchange type.
132
14.3.2. Spin Orbit Interaction as Correction to Central field Approximation
The correction of the approximate results obtained from the central field. Two
terms are omitted in the central field approximation: the difference between the
actual electrostatic interaction between electrons and the average interaction that is
included in the central field, and the spin orbit energy. The latte is an interaction
energy between the spin and the orbital motion of each electron and has the form
u r u r
x (r
k
k )Lk .Sk (1)
u r u r r
here, L k is the orbital angular momentum operator rk x k ofpthe kth electron and has
u r u2r
the properties of the J operator. The eigenvalues of L k and Lkz are given in terms of
the quantum numbers l and ml for the electron as l ( +l1 h)2 and ml h , respectively.
r 1
Sk is the spin angular momentum hs k of the kth electron. The function x ( r is
) given
2
by
1 1d V
x (r =
)
2m 2 c2 r d r
in terms of central field potential V (r).
In considering the effect of these terms, we shall assume that the perturbed
eigenfucntions, which are linear combinations of various configuration wave
functions, have only negligibly small amounts of all but one configuration mixed in
them. From perturbation theory it is apparent that this is the case if the inter
configuration matrix elements of the perturbation are small in comparison with the
energy intervals between unperturbed configuration energies.
It can be shown that from equation (1) that includes electron in full shells is
zero, since the function x ( r is
) the same for all electrons in the shell and contributions
from electrons with opposite ml and ms cancel. Thus the electrons in the full shell can
be ignored and the summation extended only over the remaining electrons. The case
in which there is just one electron outside the full shells is of interest in connection
with the ground state and low excited states of alkali atoms.
133
14.4. LET US SUM UP

In this lesson we have applied the scattering theory to atomic and molecular
structures. Also we discussed the couplings schemes and doublet separation in alkali
atoms in detail. We also presented Hydrogen Molecule problem , Heitler and
London Method and Spin Orbit Interaction .
Check your progress
1. Briefly explain the coupling schemes.

2. Write down Hartree Fock equation.
3. Explain the term spin orbit interaction.
4. Discuss the hydrogen molecular problem.
5. Explain the Heitler and London method.
6. Discuss the Spin orbit coupling. Explain how it can be treated as a Correction
to central field approximation.
14.6 References

Publishers.
Publishing house.
134
LESSON 15
SEMI CLASSICAL THEORY OF RADIATION
AND QUANTUM FIELD THEORY

In this lesson, we are going to discuss the interaction atoms with radiation
field. We are going to discuss spontaneous emission, induced emission , induced
absorption. We are going to define the Einsteins coefficients for the above three
process. Also we will be deriving an expression for transition probability. Finally we
will be presenting short notes on selection rules and forbidden transition.
15.2. Introduction
Semi classical theory is a mixture of classical & quantum picture matter i.e.
atoms are assumed to exist in discrete states. But the electromagnetic radiation with
which it interacts is described classically. The semi classical theory is insufficient to
describe the spontaneous emission of radiation because the electromagnetic field
described classically.
15.3. Interaction of Atoms with Radiation Field

An atomic system in the absence of external radiation will exist in a stationary
states. However in the presence of external radiation there will be both absorption and
emission. The atomic system may attain in thermal equilibrium with a external
radiation. This may happen in 3 ways namely i) induced absorption ii) spontaneous
emission and iii) stimulated or induced emission.
15.4. Induced Absorption

Let us suppose an atom is an low energy state E1 by absorption of quantum of
radiation it is raised to higher state E2 , then we have E2 - E1 = h w ,where w - angular
frequency of the photon. This process is known as induced absorption. The
probability of occurrence per unit time of this absorption transition E1 to E2 depends
on the properties of the state E1 and E2 and it is proportional to the rate at which the
photon fall on the atom and therefore to the spectral energy density.
135
E2
E2 - E1 = h w
E1
15.5. Spontaneous Emission
Let us suppose that the atom is in the highest state E2 .Usually the life time of
the exited state is very small, so the atom jumps to the lower energy state E1 and it
emits the photon of frequency u (angular frequency w ) / E2 - E1 = h w . This process
is known as spontaneous emission. Its probability is determined by the properties of
the state E2 & E1
E2
E2 - E1 = h w
E1
15.6. Simulated or Induced Emission
An atom is an excited energy state E2 by the influence of electromagnetic field
of a photon can come to the lower state E1 by emitting an additional photon of same
frequency. This is known as stimulated or induced emission and its probability
proportional to the properties of the state E2 & E1 and to the spectral energy density.
E2
E 2 - E1 = h w
E1
15.7. Einsteins co-efficients for Induced Absorption and

Spontaneous Emission, Induced Emission
N2
E2
E 2 - E1 = h w
1 N1 E1
136
The above figure represents the two energy states of an atomic system
corresponding to the energy E1 and E2 . Let N1 and N2 be the number of atoms per unit
volume in the energy level 1& 2. Atom in the lower energy level can absorb radiation
and get excited to the energy level 1 and 2. The excitation process occurs on by in the
presence of radiation. Such a process is termed as induced absorption or stimulated
absorption. The rate of absorption depend upon the energy density u( w ) associated
with the radiation field corresponding to the frequency.
E1 - E2
w =
h
The energy density u( w ) is defined in such a way that u( w ) d( w ) represent
the number of atoms per unit volume with in the interval w and w +d w . Thus the
number of absorption per unit time per unit volume is proportional to N1 u( w ).
N1 u( w ) = B12 N1 u( w )
Where B12 constant of proportionality and it is called Einsteins co-efficient
for induced absorption.
On the other hand when an atom, is an excited state, can make transition to the
lower energy state through the emission of electromagnetic radiation. However the
absorption process, emission process can occur two different ways.
The 1st refers to the spontaneous emission, in which the atom in the excited
state emits radiation even in the absence of any incident radiation. It is thus not
stimulated by any incident signal but occurs spontaneously. The rate of spontaneous
emission is proportional to the number of atoms in the excited state. If we represent
the co-efficient of proportionality by A21 then N2 A21 would represents number of
spontaneous per unit volume per unit time to the lower energy level.
The 2 nd refers to the stimulated emission in which an incident signal of
appropriate frequency triggers an atom is an excited state to emit radiation. The rate of
transition to the lower energy level is directly proportional to the energy density of the
incident radiation at the frequency w . Thus the number of stimulated emission per unit
volume per unit time is given by N2 B21 u( w ) . The quantities A21, B12, B21 are known
as Einstein co-efficients and are determined by the atomic system.
137
15.8. Relation between Einstein Co-efficients

At thermal equilibrium the number of upper ward transition must be equal to
the number of downward transition. Hence we may write
N1 B12 u( w ) = N2 A21 + N2 B21 u( w )
[N1 B12 - N2 B21] u( w ) = N2 A21
N 2 2A 1
u (w )=
1 -
N1 B 2 N2 2B 1
N 2 A21
=
N
N 2 1 B12 - B21
N2
Or A2 1 (1)
u (w )=
N1
N B1 -2 B2 1
2
From Boltzmanns law we have the following expression for the ratio of the
population of the 2nd level at temperature T
( E2 - E1 )
N1 KB T
=e
N2
N1 hw
= e KB T
N2
where KB Boltzmanns constant. Equation (1) becomes
A2 1
u (w )= hw
KB T
e B 1 -2 B2 1
For most of the time B12 = B21

A2 1
u (w )= (2)
B2 (e
1
hw
KB T
-1 )
According Planks law, the energy density of the radiation is given by
hw 3 1
u (w =) 2 3. hw
p c e KB T - 1
Using the above relation equation (2) becomes

138
hw 3 1 A2 1
. hw =
2 3
p c e KB T - 1
hw
(
B2 1 e K B T - 1 )
A2 1 hw 3
=
B2 1 pc2 3
hw 3
A= B 2 3
p c
15.9. Expression for Transition Probability
The Hamiltonian of the system consisting of an atom in the presence of the

radiation field is given by
p2 1 2 e
2 u er u r u r u r
2m p m8 c 2
(
H = V + E + H d +t p 2- A . + A . +2 p
m 2 c
) ( ) 2
A (1)
H = H1 +H2 +Hinteraction (2)

P2
Where H1 = + V for the atom
2m
1
H2 =
8p
( )
E 2 + H 2 dt for the radiation field
-e u r u r u r u r
Hint =
2m c
(
p. A+ A )
. p the interaction of the Hamiltonian.
e2
A 2 can be neglected because the vector potential A is very small.
2mc 2
urururur
Now we are going to prove that p. A= A ( ) ( )
. p
u r u r u r u r
In general ( p. )Ay = A.( yp)
u r u r u r u r
( p. )yA = -.ih ( yA )
u r u r u r u r
= ih-( A. )y +A. (y
)

urur
For coulomb gauge . A =0
urur
= ih- A. y
( )

uruuuuur
(
= A - hi y )
139
urur
( )
= A. py
-e u r u r-ue r u r u r u r u r u r
i.e.
2m
(
p. +A A. =p A
c 2m c
)
. +p .A p ( )
The interaction of Hamiltonian becomes
-e u r-eu r u r u r
(
= p. A . p
2A
2m c m c
) ( )
u r
Now the vector potential A can be written as
u r r r
A = $e 0 cA o-k s r(w . t )
Where e$ is the unit vector along the direction of propagation which is perpendicular
u r
to the vector potential A .
Now expressing the vector potential in terms of the exponential function
r r r r
u r ei (
k -.wr )t - i ( wk-. r )t
e+
A = e$ 0A
2
From time dependent perturbation theory the probability amplitude for transition from
an initial state n to final state s is given by
1 iws nt
cs n=s V ni n t e d t
ih
En - Es
where ws n=
h
Now,
urur
i n t n =s A . p n
s V
-e urur
= s A. p n
2m c
r r r r
-e $ A0
= e s + ei . k e r-. ip +k nr
2m c2
( - iw t
e ewi t
)
On solving the above equations the transition probability for emission become
2 w + ws n
2 2 r r
s i n t
e A $
2
2
cs n = s e p 0
n. e
i .k r
4m ch 2 2 w + ws n
2
2

140
r r
For atomic dimension the value of ei .k r
1 . So the transition probability for emission
becomes
w + ws n
2 2 s i 2 n t
e A
0 e$
2
2
cs n = s p n
4m ch
2 2 w + ws n
2
15.10. Electric Dipole Transition
The transition probability for emission is given by
2 w + ws n
2 2 r r
s i n t
e A $
2
2
cs n = s e 0
p i .k r
n. e
4m ch
2 2 w + ws n
2
2

If we consider the atomic dimension the atomic radius is very small. So we can
expand the exponential form as
r r r r2
r r i
ei .k r= 1 .+ .
( ) (+.-i. k. )+r.
.k r
. .
1 ! 2 !
The magnitude of the successive term are
2 3
2pr 1 2pr 1 2pr
1: : :
l 2 l 6 l
r r
The magnitude of the successive term decreases by the factors of but is the
l l
r r
order of 10-3 cm. So in this case we can approximate ei .k r
=1.
2 w + w
s n
2 2 s i n t
e A
0 e$
2
2
cs n = s p n
4m ch
2 2 w + ws n
2
2

Transition for which the probability can be computed using the above
approximation is called electric dipole transition. Since only the matrix element of the
r
electric dipole moment e$ . of
r the particle is involved.
141
15.11. Forbidden Transition

r
It may happen that the matrix element s r n is zero for particular state
r r
between S and n. In that case the approximation ei .k r
;- 1is
1 not valid. But it should be
expanded in the series of spherical harmonics.
r r r r r r r r
( )
ei .k jr= k0 .r 3i+
1 j. ( ) ( )
kc 1 ro- psq2 j5 k. 2 rcq po s
r r
where q is the angle between k and r
The dominant factor in the nth term of the above series is proportional to the
r rn r r
( )
k . r , if k . <<1.
r Thus if the dipole matrix element vanishes, but the next term of
each series does not vanish. The transition matrix element is reduced by a factor that
has the magnitude of the order of Ka where the linear dimension of the particle wave
function of are of order of a. The transition of this type is called forbidden transition,
since a probability is reduced by a factor (Ka)2 with respect to dipole or allowed
transition.
15. 12 LET US SUM UP
In this lesson, we have studied the interaction atoms with radiation field. We
analysed spontaneous emission, induced emission , induced absorption. We
determined the Einsteins coefficients for the above three process. Also we obtained
an expression for transition probability. Finally we presented short notes on selection
rules and forbidden transition.

Check your progress
1. Define density operator and density matrix.
2. How will you distinguish spontaneous and stimulated emission of
radiation.
3. Discuss the interaction of atom with electromagnetic field and hence show
that probability per unit time for induced emission and absorption are
equal.
4. Distinguish induced and spontaneous emission. Define Einsteins A and B
coefficients and obtain the relation between them. Write down the
142
expression for the transition probability per unit time for spontaneous
emission.
5. Discuss in detail about the forbidden transition
6. Explain the selection rules.
15.14 References
Publishing house.
143
LESSON 16
QUANTUM FIELD THEORY
16.1 AIMS AND OBJECTIVES
In this lesson,we are going to studied about quantization of wave fields. We

will be discussing the Legrangian , classical Hamiltonian equations and non
relativistic Schrodingers equations. Also we will be presenting creation , destruction
and number operators and anti Commutation relations.
16.2 . Quantization of Wave Fields: Introduction
A wave field is specified by its amplitudes at all points of space and the
dependence of these amplitudes on the time, the same way as system of particulars is
specified by the positional co-ordinates qi and their dependence on the time. The field
has an infinite number of degrees of freedom and to use the amplitudes y (r,t) at all
points r as co-ordinates is analogous to a system that consists of an infinite number of
particles and with the particles co-ordinates qi(t) respectively.
y can be expanded in some complete ortho normal set of functions uk :
y (r,t) = Sk ak (t) uk (r) (1)
where ak is regarded as the fields co-ordinates In classical particle theory, both total
and partial time derivatives were defined in connection with a function F(qi, pi, t) of
the co-ordinates, momenta and time. Also a function F(y, p, t) can depend explicitly
on the time as well as on the field, so that it is important to distinguish between dF/dt
and F/dt.
16.3. Classical Lagrangian Equation

The Lagrangian L(q i, q i, t)is a function of the time and a functional of the
possible paths qi(t) of the system. The actual paths are derived from the variation
principle.
144
t2
d L. dt = 0, dqi(t1 ) = dqi(t2 ) = 0 (1)

t1
we expect the field Lagrangian to be a functional of the field amplitude y(r,t). It can
be expressed as the integral over all space of a Lagrangian density L :
.
L = L (y , Dy ,y , t )d 3 r (2)
where y is y/t or dy/ d t
The appearance of y as an argument of L is a result of the continuous dependence
of y on r. The variation principle that corresponds to equation (1) is
t2 t2 t2
3
d L dt = d L dt d r = (d L) dt d 3 r = 0 (3)
t1 t1 t1
where the variation dy of y is subject to the restrictions,

dy (r , t1 ) = dy (r , t 2 ) = 0 (4)
The variation of L can be written as,

u r u r u r
L L Ly
dL = d
+y d + d y & (5)
y z y y
x y x &

x
where the summation over x, y, z implies the sum of three terms with y and z
y
substituted for x we have, dy = (dy ) , d = (dy )
t x x
Then equation (3) becomes
u r u r u r
t0 L L L
d + y d . +y( ) d y(0= ) d t 3d r (6)
y xy z y y x t&
t1
x
Then summation terms in equation (6) can be integrated by parts with respect to space
co-ordinates; surface terms vanish either because y falls rapidly at infinite distance or
because y obeys periodic boundary conditions. The last term in equation (6) can be
integrated by parts with respect to time and the boundary terms vanish because of
equation (4). Now equation (6) can be written as,
145
u
uL r r u r
t0
L L 3
- - = d yd t d r0 (7)
t1 y x y z x y
y t &
x

since equation (3) is valid for an arbitrary variation dy at each point in space,
equation (7) is equivalent to the differential equation,
u r u r u r
L L L .
- - = 0 (8)
y x yz x yy t &
x
Equation (8) is the classical field equation derived from the Lagrangian density
L (y,y,y,t).
.
16.4. Functional Derivative
In order to peruse further the analogy with particle mechanics, it is

desirable to rewrite equation (8) in terms of L rather then L . For this we require
derivatives of L with respect to y a n d y at particular points. These are called
L L
functional derivativesand are denoted by and
y& y
Expressions for them can be obtained by dividing up all space into small cells and
replacing volume integrals by summate over these cells. The expressions for the
function derivatives are as follows.
u r u r
L
L Ld L L
.
= L t = -
d 0 d y d t y

xy z x y
(9)
yy r j
j j

x
Here all the dyi and dy i are zero except for particular dy j, and dti is the volume of
the cell. Similarly, the functional derivative of L w.r.t y is defined by setting all the
dy i and dy i zero except for dy i
u r
L L d L
= L t = (10)
y ddr j 0 y & jd j t& y
146
Here again at point r at which the functional derivative is evaluated is in the jth cell.
Substituting equations (9) & (10) in (8) gives,
L L
- = 0 (11)
t y& y
which closely resembles the Legrangian equations
d L L
- = 0 for a system of particle.
d t &i q i q
16.5. Classical Hamiltonian Equations
The momentum canonically conjugate to y j can be defined as in particle

mechanics to be the ratio of dL to the infinitesimal change dy j when all the other dy
. j
and all dy i are zero we thus obtain.
dL L
Pj = = dt j (1)
dy j y j
L
It follows that Pj = dt j (2)
y j
L
The Hamiltonian H = Pi y i - L = y i dt
. i -L (3)
i i y i
H is written as the volume integral of a Hamiltonian density H and assume that the
cells are small enough so that the difference between a volume integral and the
corresponding cell summation can be ignored; we than have,
L L
H = H d 3r H = py -L p = (4)
y y
The true field Hamiltonian H given in equation (4) which is a functional of y and p
from which y has . been eliminated.
. The classical Hamiltonian equations of motion
will be derived without further recourse to the cell a approximation. The variation of
L produced by variations of y and y can be written using equation (4) and
L / L /
- = 0
t y& / y /
L / L /
d L =d y dd+ y . p&r = d3 y d( +&p& 3
d ry )
y / y /&
147
= [ d (p+&y) p &d- y& y d p] d 3 r
(p& d y- y& d )pd

3
= d H + d L+ r (5)
The variation of H produced by corresponding variations of y and p can be written,

H / H /
d H = d y
+ d p . d3 r (6)
y / p /
It follows from the earlier discussion of functional derivatives that.
H /
H H
= -
y / y x y z
x
y

x
(7)
H /
H H
= -
p / p x yz x p

x
Comparison of equations (5) & (6) for arbitrary variations dy and dp then gives the
classical field equations is Hamiltonian form.
H / H /
y& = =
p &- (8)
p / p /
The Hamiltonian equation for the time rate of change of a functional F of y and p
can now be found. We express F as volume integral of the corresponding functional
density F (y , p , t ) , which for simplicity is assumed not to depend explicitly on time
or on gradients of y or p . Hence,
u r u r
d
F F F F
= + +& p & d 3 r
y
d t t y p
F F/ H/ F/ H /
= + - d3 r
t / y / p p/ y/
F
= + {F, H} (9)
t
The above equation also serves to define poisson bracket expression for two function
as of the field variables. The RHS of equation (9) is not changed if F also depends on
y or p. It is clear from equation (9) that H is a constant of the motion if it does not
depend explicitly on time; in this case H is the total energy of the field.
148
16.6. Field Quantization of the non Relativistic Schrdinger

Equation
Consider the Schrdinger wave equation,

y - h 2 2
ih =V ry + (ty , ) (1)
t 2m
Schrdinger equation. For this reason, As a first example of the application of the
field quantization technique, we consider here the quantization of the non relativistic
Schrdinger equation. The application implies treating (1) as though it were a
classical equation that describes the motion of some kind of material fluid. The
resulting quantized field theory is equivalent to a many particle field quantization is
often called second quantization, and the transaction from classical particles
mechanics to equation (1) constitutes the first quantization.
Classical Lagrangian and Hamiltonian equations,
The Lagrangian density may be taken to be
u r h2
&
L = h y i-* y*V y
. (r y -, y
t) y
* (2)
2m
y and y * can be varied separately to obtain the Lagrangian equations of motion.
The equation of the form that results from variation of y is
h2
- ih y&=* V-* y (2r +, yt ) *
2m
Which is the complex conjugate of (a) variation of y * gives
h2
ih y& = V - y 2r + y
(t , ) (3)
2m
The momentum canonically conjugate to y is
u r
L
p = =y i h * (4)
y&
However y * does not appear in the Lagrangian density so that p is identically zero. It
is therefore impossible to satisfy the corresponding classical poission bracket relation,
so that y *, p cannot be regarded as a pair of canonically conjugate variables. p
is the second momentum.
The Hamiltonian density is,
149
u u r u r -i h -i
H = p &- yL= p y
.V p y (5)
2m h
The Hamiltonian equations of motion obtained from the classical field equations with
the help of functional derivates are,
i hi
y& = + y 2
-Vy
h
2m
i i h 2
p& = Vp- p
h 2m
We have thus shown, from the point of view of classical field theory, that the
Lagrangian density and the canonical variables and hamiltonian derived from it are in
agreement with the wave equation.
16.7. Creation, Destruction, and Number Operations

Consider the commutation relations
ak ( t )a ,l t =(*u )rk ' u( r)-3l r(d r)3 dk( =r 3' dd )l r ' (1)
for the operations ak and ak + were solved in connection with the harmonic oscillator. It
was found that the solution of aa+- a +a=1 in the representation in which a+a i s
diagonal, consists of the matrices.
0 0 0 0
0 1 0 0 1
1 0 0 0
a = 0 0 2 2 0 a+ = (2)
0
1
2
1 0 2 0
0 0 0 3 2 1
2
0 0 3 0
Let N be Hermitian, its eigenvalues are real.
+
N= Nk
k where N k = ak ak (3)
The states of the quantized field, in the representation in which each Nk is diagonal,
are the kets
n1 , . .2. . . .n. . . . . . . .n. . . .k. , . . . . . . . . . . . . . (4)
Where each nk is an eigen value of Nk and must be a +ve integer or zero. We also
have the relations.
ak 1 , . . . n. . . . . .k . . . n. . . . 1. . , . . n.k . . =. . . 1. ./ .2 .n. . . ,k..............1,........................
. . .-. n
1 / 2 (5)
ak 1 , . . . .n. . . . .k . . . .n. . . . 1. , . . .n =k. (. . . . . .). .n.+. . 1k ., ..... . n. . +. . . . . . . . 1 , . . . . . . . . . . . . . . .
150
These ak + and ak are called creation and destruction operators for the state k of the
field.
The number operation Nk need not be a constant of the motion, that
N = Nk is a constant The rate of change of Nk is given by,
ih &Nk = ak
ak , H
h2
where H = a j la u *j V* j
ul . u+l u3 d r
j l 2m
-h 2 2
= a j l aj *
u Vl u+
3
d r (6)
j l 2m
diagonal, Nk is constant only if all the volume integrals in equation(6) are zero for
which either j or l is equal to k. These integrals are just matrix elements of the one
particle Hamiltonian. The integrals in equation (6) are than E l d jl , and the field
Hamiltonian becomes,
+
H= ak
k ak Ek = N k Ek
k
This particular N representation is the one in which H is also the ket

n1 , ...... nk , ...... > has the eigen value n
k
k E k for the total energy operator H. It is
apparent that all the Nk are constant in this case.
16.8. Anti Commutation Relations

The commutation relations are
[y (r ) y. r( ] ='y) r+ ( y) + .r= ( ' ) 0

(1)
r y+r r d= 3
y (),(') r ( )
It is necessary to modify the above equations if we are obtain a theory of particles that
obey the exclusion principle. It is reasonable to require that this modification be
made in such a way that The quantum equation of motion of y is the wave equation,
h2
ih y& = V - y 2r - (yt , )
2m
when the Hamiltonian has the form,
151
h2
H = y + .y ++y
V yd 3
r
2m
It was found by Jordon and Wigner that the descried modification consists in
the replacement of the commutator brackets,
[A,B] AB BA
By anticommutator brackets
[A,B]+ AB + BA
This means that equations (1) are replaced by,
[y (r ) y, r( ]+ 'y=) r ( y) r( +'y ) r ( y' =r) ( ) 0

y + (r ) y+, r( y + +
' =) r ( y) r( + y+ +
' ) r ( y' =)r ( ) 0 (2)
+

y (r ) y+, r ( 'y=r)+ ( y)r +(+ 'yr ) (=y r ' ) r(d - )3 (r ' )
+
It then follows that

[ak , aL]+ = ak al + al ak = 0
[ak +, aL+]+ = ak +al+ + al+ ak + = 0
[ak , aL+]+ = ak al+ + al+ ak = d kl (3)
We define Nk = ak + ak as before and notice first that each Nk commutes with all the
others, so that they can be diagonalized simultaneously. The eigen values of Nk can be
obtained from the matrix equation,
Nk2 = ak + ak ak + ak = ak + (1-ak + ak ) ak = ak + ak = Nk (4)
1 11 2
If Nk is in diagonal form and has eigen values nk , nk .........., it is clear that Nk is also
in diagonal form and has eigen values nk 12 , nk 112, ... Thus the matrix equation is
equivalent to the algebraic equations.
1 11 2
2 1 11
nk = nk nk = nk ...........
For the eigen values. These are quadratic equations that have two roots 0 and 1. Thus
the eigen values of each Nk are 0 and 1, and the particles obey the exclusion principle.
The eigen values of N = N k are the +ve integers and zero as before. The earlier
expressions for the Hamiltonian in creation, destruction and number operators are
unchanged, and the energy eigen values are n E k k .
152
16.9 LET US SUM UP
In this lesson, we have discussed about quantization of wave fields. We

studied the Legrangian , classical Hamiltonian equations and non relativistic
Schrodingers equations, in detail. Also we presented creation , destruction and
number operators and anti Commutation relations.
Check your progress

1. Define density operator and density matrix.
2. Obtain the Legrangian equations of motion.
3. Obtain the classical Hamiltonian equation.
4. Arrive at the non relativistic Schrodingers equation.
5. Explain the creations, destruction and number operators.
16.11 References
Publishing house.
153
M.Sc. DEGREE EXAMINATION, APRIL 2008

Model Question
Physics
QUANTUM MECHANICS
Time : Three hours Maximum : 100 marks
Answer any five questions

Each question carries 20 marks. (5 x 20 = 100)
spectrum. Write down the physical meaning of matrix elements.
2. Explain the main stages involved in the WKB approximation. Use the
technique to obtain the asymptotic solution of the one dimensional
Schrdinger equation. Also arrive at the solution near a turning point.
3. Develop the relativistic wave equation for the particle of spin zero. Discuss the
difficulties involved in interpreting the probability density. Derive the
expression charge and current densities from the continuity equation.
4. Prove the following properties of Dirac matrices on the basis of Diracs
theory :
a) Square of matrices are unity
b) Eigenvalues of the matrices are 1.
c) Trace of the matrices is zero.
d) Dimension of the matrices are of even order.
e) The matrices anticommuting among themselves.
5. Obtain an expression for the Born approximation amplitude and hence derive
the Rutherford formula for Coulomb scattering. Enumerate the two conditions
for the validity of born approximation.
6. Discuss the Thomas Fermi Model and Hartrees Self Consistent Model in
detail. Briefly explain Heitler and London method.
7. Distinguish induced and spontaneous emission. Define Einsteins A and B
coefficients and obtain the relation between them. Write down the expression
for the transition probability per unit time for spontaneous emission.
8. Obtain classical Lagrangian equation. Write notes on Creation, Destruction
and Number Operators. Also discuss their anti commutation relations.
154

Model Question
Physics
QUANTUM MECHANICS

1. What do you understand by equation of motion? Obtain the equation of motion

in the Heisenberg picture and state the distinction between Schrdinger picture
and Heisenberg picture.
spectrum. Discuss the significance of zero point energy.
3. Obtain Perturbation theory for non degenerate states. Show that the first order
perturbation energy for a non degenerate state of a system is just the
perturbation for average over the corresponding unperturbed state of the
system. Distinguish stationary and time dependent perturbation theory.
4. Explain how the ground state energy could be evaluated using the variation
method and hence find the ground state energy of hydrogen atom.
5. Obtain the spectrum of eigenvalues for J2 and Jz. Hence calculate the angular
momentum matrices for j = . What are C.G. Coeffifiient? Briefly explain
their significance.
6. Develop Diracs relativistic wave equation for a free particle and obtain its
plain wave solution. Give the significance of negative energy states. What are
the consequences of Diracs interpretation?
7. What do you meant by partial wave analysis? Employ this method to derive an
expression for the scattering amplitude in terms of phase shift. Obtain an
expression for scattering cross section for Coulomb potential.
8. Discuss the interaction of atom with electromagnetic field and hence show
that probability per unit time for induced emission and absorption are equal.
Obtain classical Hamiltonian equation.
155

Model Question
Physics
QUANTUM MECHANICS

1. What are the conditions under which an operator can be represented by a

diagonal matrix. Write down the physical meaning of matrix elements.
Explain Kronig penny model.
2. There is no first order stark effect in the ground state of an atom Illustrate.
Define the degeneracy of an energy eigenvalue and illustrate its removal by
considering the stark effect in the n = 2 level of the hydrogen atom.
3. Obtain the expressions for the first order corrections to the energy and
wavefunction making the use of the non-degenerate stationary perturbation
theory. Obtain an expression for harmonic perturbation.
4. Give the quantum mechanical definition of angular momentum. Find the
angular momentum matrices for j = 1. List the symmetry properties of C. G.
coefficient.
5. Show that the angular momentum of electron is a constant of motion only if
we include the spin. Establish the Diracs relativistic wave equation for an
electron and calculate its magnetic moment.
6. Define the terms scattering amplitude and phase shift. Obtain an
expression for scattering amplitude in terms of phase shift. State and explain
optical theorem. Discuss in detail the problem of Yukawa potential.
7. Explain the terms doublet separation and doublet intensity. Explain how the
spin orbit interaction can be accounted as correction to central field
approximation.
8. Distinguish induced and spontaneous emission. Define Einsteins A and
B coefficients and obtain the relation between them. Write down the
expression for the transition probability per unit time for spontaneous
emission.

Physics Quantum

Diunggah oleh

Informasi Dokumen

Hak Cipta

Format Tersedia

Bagikan dokumen Ini

Bagikan atau Tanam Dokumen

Opsi Berbagi

Apakah menurut Anda dokumen ini bermanfaat?

Apakah konten ini tidak pantas?

Hak Cipta:

Format Tersedia

Physics Quantum

Diunggah oleh

Hak Cipta:

Format Tersedia

This watermark does not appear in the registered version - http://www.clicktoconvert.

LESSION 1: INTRODUCTION TO QUANTUM

1.0. Aims and Objectives:

In this unit, we are going to discuss

1.1. Quantum Mechanics: An Introduction

1.2. Difficulties with Classical Physics

Classical Physics deals primarily with macroscopic phenomena. Most of the

1.3. The Birth of Quantum Mechanics

The early development of atomic theory consisted of efforts to overcome these

Quantum Mechanics can be regarded as the fundamental theory of atomic

1.4. State Vectors

1.5. Hilbert Space

It is an abstract space. The vector spaces of quantum mechanics are called

1.6. Dirac Notation

Dirac introudced the symbol y to denote an abstract state vector distinct

Corresponding to every vector y there is a conjugate vector y for which

Dirac used to notation y .

1.7. Norm and Scalar Product

The length or norm of the vector y is denoted by y y . By definition

y y is a real non-negative number.

complex) number which is defined to be the scalar product of f and y , taken in

In the Schrdinger picture, f y f* ( x )yx( d)t . It is to be noted that

The scalar product of two vectors is linear in the second vector.

f ( c y + cc' )=f c y' + c f c (2c)

1.8. Dynamical variables as Operators

Each dynamical variable A (x,p) is represented in quantum mechanics by a

system. The effect of an operator A on a wave function is to convert it into another

A ( 1yc +1 c 2y) =c2 ( A1y + )1 c 2y( A 2 )

1.9. Change of Basis

If the vectors e i are on orthonormal basis for a N dimensional space,

Then a new basis can be constructed by forming N linear combinations of these

If Equation (2) is substituted the above condition becomes

u i u*k ( ekj , lle )=i d j (4)

Using equation (1) we get

In other words, the matrix U = ui j must satisfy

It follows that, U must be non singular and

{e } .If a vector x is represented with respect to the basis {e } by

and with respect to basis {e' } by

can be solved for xi' in terms of xi .

1.10. LET US SUM UP

In this unit, we have studied

1.11. LESSON END ACTIVITIES

6.Angular Momentum Techniques in Quantum Mechanics V. Devanathan

2.0. Aims and objectives:

2.0. Aims and Objectives:

In this lesson we are going to study

2.1. Schrdinger Picture

The subscripts refers to the ket as viewed is Schrdinger picture. T h e t o t a l

Since H = H because H is hermitian. Assuming H to be independent of time

A special case of interest will b when W s is time independent, and when W s

2.2. Heisenberg Picture

substituting for ket and bra vectors

This equation is called equation of motion in Heisenbergs picture. This is

where { F , H} is poisson bracket.

So the transition from classical to quantum mechanical picture can be made by

2.3. Interaction Picture (or Dirac Picture)

In the interaction picture, the Hamiltonian is split up into two parts as H = H0

Differentiating equation (1)

In interaction picture, part of the time dependence is assigned to the state

2.4. LET US SUM UP

2.5. LESSON END ACTIVITIES

3. Compare the equation of motion in Schrodingers picture with that of

Obtain the equation of motion for <A> in the Schrodingers picture

3.1. Aims and objectives: